Skip to main content

Full text of "The place-names of the Liverpool district; or, The history and meaning of the local and river names of South-west Lancashire and of Wirral"

See other formats







^Liverpool . . . t/tai Saxon hive.' — Matthew Arnold. 

' Liverpool . . . the greatest covunercial city in the world' 

Nathaniel Hawthorne. 

' That's a great city, and those are the lamps. It's Liverpool.' 
' Christopher Tadpole ' (A. Smith). 

' In the United Kingdom there is no city luhichfrom early days 

has inspired me with so -much interest, none which I zvould so 

gladly serve in any capacity, however humble, as the city of 


Rev. J. E. C. Welldon. 



^he l)i0torj) mxb Jttciining oi the ^oral aiib llibev 
^mncQ oi ,S0xitk-to£0t |£ancashtrc mxlb oi SEirral 



•respiciendum est ut discamus ex pr^terito. 

















DOMESDAY ENTRIES - - - - - 20 


LIVERPOOL ------- 24 


HUNDRED OF WIRRAL - - - - - 75 



This little onomasticon embodies, I believe, the first at- 
tempt to treat the etymology of the place-names of the 
Liverpool district upon a systematic basis. In various 
local and county histories endeavours have here and there 
been made to account for the origin of certain place-names, 
but such endeavours have unfortunately only too frequently 
been remarkable for anything but philological, and even 
topographical, accuracy. They are, however, generally 
chronicled, as a matter of record, in the present mono- 
graph, with such criticism and emendation as may have 
been thought necessary. 

The science of philology has made rapid strides since 
the days when Syers, in his History of Everton, solemnly 
asserted that etymology, a branch of philology, was neither 
more nor less than " guessology " ; but even to-day, after all 
the accessible historical and philological evidence bearing 
upon a name has been thoroughly sifted and carefully 
weighed, there sometimes remains an element of uncer- 
tainty that creates a hiatus which must be filled by guess- 
ing — but, still, by what Professor Skeat has called, in this 
connection, " reasonable guessing,"^ not the kind of etymo- 

^ Dr. Sweet says, in the preface to his new Anglo-Saxon Dictionary: 
'The investigator of Old Enghsh ... is often obliged to work by 
guesswork, until some one else guesses better." 


logical jumping at conclusions which has, for example, 

induced a Welshman to claim that the name Apollo is 

derived from the Cymric Ap-Jiaul, ' Son of the Sun ' ; an 

Irishman to assert that the Egyptian deity Osiris was of 

Hibernian descent, and that the name should consequently 

be written O'Siris; a Cornishman, saturated with the 

Phoenician tradition, to declare that his Honeyball is a 

corruption of Hannibal ; a Scotsman to infer an affinity 

between the Egyptian Pharaoh and the Gaelic Fergus ; and 

even an Englishman to calmly asseverate that Lambeth 

(the ' lamb-hithe '), containing the palace of the Archbishop 

of Canterbury, derived its name from the Thibetan llama, 

'high-priest,' and the Hebrew beth, ' house.' 

While, however, in England, we bring our guessing 

powers into operation only, as a rule, after the lapse of 

centuries, in America it sometimes happens that a place 

receives its name one day and the next (so to speak) the 

origin of that name is shrouded in mystery, as witness the 

following characteristic extract from a recent number of a 

Western States journal : 

" Nobody around the oilfields seems to know why the new field is 
called Chipmunk. Most aver that it has always been Chipmunk ever 
since the time of the mound-builders. Others have it that the first 
white settler was eaten by chipmunks, ever since which notable event 
a pure white chipmunk has haunted the valley, scaring other chipmunks 
to death. Chii)munk may also be called Chipmunk because there are 
no chipmunks there." 

In order to impose a more or less recognised limit upon 
the so-called district of Liverpool, it has, for the purpose of 
this treatise, been divided into two hundreds — that of West 
Derby, which comprises practically the whole of south-west 
Lancashire, and that of Wirral, which embraces the tongue 
of land separating the estuary of the Mersey from that of 
the Dee. The names enumerated in the body of the work, 
which is arranged in alphabetical form with respect to the 
two hundreds, I have summarized herewith, according to 
their linguistic origin : 

••S s. 

• (U O o 

.::5 3 






w o 
~-^ o 

■ ^JS 

^ en x; ^ 

g U3 -" 

a. V- 
> (J 


1 rt 

J5 "rt O 3 ^ 

^ "13 ti ^ -^ 


. >th ;:: -^ ^ 

^ c S 

fl rt = 






























<u u "S 

o . 




O > >^ <V V ^ 

tflj; t« en X K OJ-^^-Srlr^ ,- >- - :>_LJ "n.'^ 
J-(-^(~in^-\L-r»v- . -t-J -^ K J-" G >_i ►* ""^ r^ U J-i 



c • (« c o 

rt o 5 




"So d ^ 
.S o o 

C 3 ij 
HJ O 1-, 

Ph Ph Ek 


c '" S 

P^ Pi f^ c>5 c^ cAi c/D 

h\^ V o v 
C 3 oj aj 

1) k- 1) o 

5 O X a; 

o cxa.5j3o 3 >> 


c . 


^5 ^ . 

0) . ^ . 

. > tJoS- G 

. s "o o 5 

c 2 = t; c 
2 wc ?j t?; 

c 2 5 -y s 2 

.s/j.s.s.s 8 o 

^ i^ u > ■-• 
13! rt rt rt -C 






'-> ^ r^ 

:=; o u 

'O T3 ;s 

O . (U 

£ (U S "in . « 

3 ^ >, O t/) S 

O 3 3 

U E 

S^ rtrt'o'5c3c3rt.So033c'3.S 


























































OpLtPMO* Cm 



*-< (11 r? 


r! tl ■" 
£ 3 Ji 

c C ° • • 

^ ^ O S "k^ . T3 aj 

■ -H 3 rt rt u r^ ' 


2 ^ 

tl o 

3 J3 



^ G 

O O 

• f -*-> 

1 — H *-. X 

§2 s 


^ >-.t3 

:* -^ a 
_ C rt ^ 

, ,, ,„ ^ .^ ^ O 3 J3 ii w 3 rt rt .-, O u > rt rt OJ 





-? < 








« ^ rt 

i- 1) o 2 


0) ■ 


c a 

C <-> o 


-— ^ 

















.ti ^ 

^ C •* t^ 














c c c ^^ -2 ^ t^ 




, o 



>< o o 2: 3 

C u a) 

'^ ;- ,-, vj 'w ' ^^ 
Cj O O 4-* *J •*-» 



. c 

'p o 


§ c c 

3 '• 

1) -^ 

■ « s 

c o o 

1) • >. 


• t/i 

. ^ o 

c <u S 

5 -s o 

in 0) 53 

j:-^ P 





















rt ^^-» 




o :i o ^ " '^ i^ 


O rt 

. (/•; ^ r- " • 

?^ •- S O M ? 

S ^ • 


a, 3 o 
iT/:: o 
OJ u Q, 


^ .- ^ c O C 

^ '_j {A ,7, _rs _^ O f^ ^ -S TT, O "" ^ *-* 
— C' P n n o « aj S^.q a, a; > > 

,5^ R ,^ ,^ ►S 


The chief fact to be gathered from these lists is that 
both in south-west Lancashire and in Wirral the Anglo- 
Saxon names are about three times as numerous as those 
of Norse origin ; and it was to be expected that, in a part 
of the country which was wrested from the British at a 
later period than is assigned to the Saxon conquest of most 
of the remaining portions of what ultimately came to be 
called England, distinct traces of Celtic nomenclature should 
be met with.^ 

" Letters, like soldiers," that acute philologist Home 
Tooke once observed, " are very apt to desert and drop off 
in a long march." Of this truism the vicinity of Liverpool 
is not behind other districts in affording good illustrations. 
The Norse Otegrimele and Otringemele, as chronicled in 
Domesday, have descended to us in the attenuated form of 
Orrell ; Levetesham is now Ledsham ; Stochestede, Toxteth ; 
Chenulveslei, Knowsley ; Herleshala, Halsall ; and so on. 
On the other hand, there are names which, in the course 
of time, have added a trifle to their length. Oxton at one 
period was Oxon ; Speke was Spec. 

Luckily we are not blessed — or the reverse — with many 
names of the " funny " order, or even of that American 
genus against which Matthew Arnold declaimed : " When 
our race has built Bold Street, Liverpool, and pronounced 
it very good," caustically observed the author of T/ie Study 
of Celtic Literahire (p. 175), " it hurries across the Atlantic 
and builds Nashville, and Jacksonville, and Milledgeville, 

1 \yords\vorth, whose Poems on the Naming of Places will be 
familiar to the reader, has well expressed in verse the changes wrought 
in Britain by the Saxon Conquest — 

" Another language spreads from coast to coast ; 
Only perchance some melancholy stream 
And some indignant hills old names preserve, 
When laws, and creeds, and people all are lost !" — 

Monastery of Old Bangor. 


and thinks it is fulfilling the designs of Providence in an 
incomparable manner." 

Possibly Arnold may have borne in mind, too, the 
ludicrous origin of such American place-names as Elberon 
(L. B. Brown), Carasaljo (Carrie, Sally, and Joe), Eltopia 
(Hell-to-pay), and Nameless, which last-mentioned town 
received its incongruous designation because a lazy postal 
official at Washington, having repeatedly been urged by the 
inhabitants of a new and thriving village in Laurens County, 
Georgia, to select a name for it, at last testily telegraphed 
to the astonished settlers, " Let it remain nameless " ; and 
Nameless accordingly the place has been ever since. But, 
as I have remarked, we of the Liverpool district have very 
few of the kind of names at which an American once poked 
revengeful fun in a set of verses, one of which ran : 

"At Scrooby and at Gonexby, 
At Wigton and at Smeeth, 
At Bottesford and Runcorn, 
I need not grit my teeth ; 
At Swineshead and at Crummock, 
At Sibsey and Spithead, 
Stoke Pogis and Wolsoken, 
I will not wish me dead." 

Still we have to confess, with some degree of sadness, to 
a Puddington, a Mollington, a Noctorum, and a Greasby, 
all in Wirral ; while, as to the Lancashire side of the 
Mersey, we have often seen and heard prettier and more 
euphonious names than, say, Ravensmeols, Bold, Maghull, 
Skelmersdale, and Chowbent. Even a silver-tongued White- 
field, who was reputed to be able to " pronounce ' Mesopo- 
tamia' so as to make a congregation weep," would, we 
should imagine, have experienced some difficulty in invest- 
ing with charm the utterance of the names which we have 
just enumerated. 


I have compiled (with special reference to the Hundreds 
of West Derby and Wirral) and appended to the Introduc- 
tion a brief glossary of some of the most frequently occurring 
English place-name components. It might perhaps be con- 
sidered amply sufficient when we recollect the Elizabethan 
Verstegan's oft misquoted couplet : 

" In ford, in ham, in ley, and tun, 
The most of English surnames [place-names] run." 

{Restitution of Decayed Intelligence in Antiquities. ) 

But we must not forget that, as Kemble says {Cod. Dipl. 
iii. XV.), speaking of Anglo-Saxon place-nomenclature : 
•' The distinctions between even the slightest differences in 
the face of the country are marked with a richness and 
accuracy of language which will surprise ..." The com- 
ponents embodied in another well-worn distich : 

" By tre, pol, and pen, 
Ye shall know the Cornishmen," 

do not, of course, come within the scope of the glossary. 

Mention should be made of the difficulties which the 
Norman Conquest was the means of strewing, like caltrops 
or chevaux-de-frise, in the path of the investigator of Anglo- 
Saxon nomenclature. "The Normans," says Skeat, "spelt 
Anglo-Saxon names anyhow." "The Normans," Kemble 
grumbled, referring to the terms in Anglo-Saxon charters, 
"could not even spell the words." "The loose manner 
of spelling the names of English places in Doomsday Book 
cannot," observed Gregson in his Portfolio of Fragmetits, 
" be wondered at when it is considered that the Normans 
had the chief hand in compiling the returns." 

Hardy, in his Introduction to the Close Rolls, portions 
of which essay have been borrowed by the writers of the 


Pipe Roll Society's introductory volume as being also 
applicable to the rolls with which they were dealing, re- 
marks : " Great ambiguity prevails in the proper names of 
persons and places which occur on the Close Rolls ; for 
these were either Latinized or Gallicized, whenever it was 
possible to do so, according to the fancy of the scribe or 
the degree of knowledge which he happened to possess. 
Thus he rendered into Latin or French a Norman or 
Saxon appellation just as he happened to prefer the one 
to the other. . . . Whitchurch is sometimes written 
De Albo Monasterio, sometimes Blancmuster or Blaunc- 

Mr. G. Grazebrook, in a paper read before the Society of 
Antiquaries in February, 1897, on the spelling of mediaeval 
names, submitted a list of sixty-one various forms in which 
the name Grazebrook is found from the year 1200. 

A good instance of the vagaries of a Norman writer or 
copyist of Anglo-Saxon is supplied in a MS. of ' The 
Proverbs of Alfred,' upon which Professor Skeat read a 
paper in May, 1897, at a meeting of the Philological Society 
{Athenccum, May 15, '97). The Norman scribe who wrote 
out this MS. has, amongst other blunders and peculiarities, 
t, and occasionally d, for final th, and, conversely, th for 
the English final t ; st (with long s) for the final ^A/; s for 
sh^ as sal for shal ; w for wh, as wat for what ; cherril for 
cJmrl ; arren for am; welethe for welt he ; chil for child; 
wen for went ; kinc for king ; wrsipe for tvorship ; hujit for 
htind, and ant for a7id. And when we find, to carry 
example a little further, an Anglo-Saxon Hweorfanliealh 
transformed in Domesday into Vurvenele, there is little 
ground for wonder that the pursuit of the etymology of a 
local name should sometimes be a tedious operation, 
although it must be said that there is scarcely need to 


spend the leisure of thirty years in endeavouring to ascer- 
tain the origin of a single place-name, as a resident of 
Kensington recently confessed in JVbfes and Queries to 
have done. 

Of course, many amateur topographical derivation hunters 
lose considerable time in persistently endeavouring to trace 
the first element of an English place-name to some physical 
feature or characteristic, simple or complex, when all the 
time the prefix is often merely a personal name, possibly 
somewhat corrupted. It has probably been the fashion in 
all ages, and in all countries, for personal nomenclature to 
be used, in greater or less degree, in place designation : 
" And Cain . . . builded a city, and called the name of 
the city after the name of his son Enoch." — Gen. iv. 17. 
" And they called the name of the city Dan, after the name 
of Dan their father."— y?^^^^^ xviii. 29. 

For the benefit of the general reader it may not be thought 
superfluous to state that, of the two State Records of which 
the most frequent mention is made in the course of the 
monograph, the immortal Domesday Book was completed 
in A.D. 1086 ; while the Testa de Nevill, or, to give the 
Exchequer collection its full title, ' Testa de Nevill, sive 
Liber Feodorum in Curia Scaccarii,' relates ostensibly, and 
with little actual variation, to the times of Henry III. and 
Edward I. (1216 to 1307), and was compiled at the begin- 
ning of the reign of Edward III. The collection of Anglo- 
Saxon charters printed by Kemble — Codex Diplomaticus 
^vi Saxonici — is comprised in six volumes, the first of 
which was published in 1839, the last in 1848. This work 
has now, to some extent, been superseded by the Cartu- 
lariicm Saxonicuin of Mr. de Gray Birch, the chief continu- 
ator of Kemble. The History of La?icashire quoted, with 
the date, for the second volume, of 1870, is Baines's excel- 


lent compilation as edited by the late John Harland, F.S.A., 
and continued and completed by Mr. Brooke Herford; 
although where necessary reference is made to Croston's 
edition, 1888-93. Ormerod's monumental History of 
Cheshire, first published in 181 9, was reissued in a revised 
form in 1882 by Mr. Helsby. Mr. Beamont's Domesday 
Cheshire and Lancashire should be mentioned in conjunc- 
tion with Colonel James's Domesday Facsimile of those two 
counties. The Publications of the Chetham Society, the 
Record Society of Lancashire and Cheshire (which printed 
the collection of Lancashire and Cheshire documents in the 
Public Record Office edited by Mr. W. D. Selby, author of 
the vade-mecum of record searchers, Thejiibilee Date Book), 
the Historic Society of Lancashire and Cheshire, the 
Literary and Philosophical Society of Liverpool, the Lan- 
cashire and Cheshire Antiquarian Society, and the Chester 
Architectural, Archaeological, and Historic Society of course 
provide splendid raw material for the student of Lancashire 
and Cheshire place-names, although here and there a paper 
with a not unpromising title, and by a well-known scholar, 
may, upon investigation, prove to be disappointing, as, 
for instance. Dr. Latham's essay ' On the Language of 
Lancashire under the Romans,' in vol. ix. (1857) of the 
Transactions of the Lancashire and Cheshire Historic 

My indebtedness for miscellaneous information to the 
above-mentioned and various other works (including the 
valuable treatises of the Rev. Canon Taylor), and to local 
antiquaries, is duly recorded in the proper place, and 
a bibliography is appended; but I should specially men- 
tion some MS. notes which Prof. Skeat, author of the 
much-used Etymological Dictionary, kindly placed at my 


Unfortunately, the aforesaid student of predial names in 
Lancashire and Cheshire cannot avail himself of a record 
which the topographic investigator in most other English 
counties can study with profit, viz., the Rotuli Hundredorum, 
or Hundred Rolls {tevip. Hen. HI. and Edw. I.), which 
contain no extract relative to the Counties Palatine of Lan- 
caster and Chester. 


{Compiled with special reference to the Liverpool District.) 


AlK ^ 

Acre \ 




Bold, Bootle 






Carl (Eastern and 
Northern, i.e., 
Angl. and Dan.) 

Chakl (Southern) 

Chorl (Midland) 

Caster (Angl. and 





Ea ) 

Ey J 



A. -Sax. ac, oak. 


Graf (prefix)"! 
Grave |- 

Greve J 



A. -Sax. cEcer, field, acre. 

A-Sax. edst, east. 

A. -Sax. here, barley. 

A. -Sax. bold, botl, dwelling. 

(i) A. -Sax. biirh, burg, Scand. borg, fortified 

place, castle, city ; (2) A. -Sax. beorh, beorg, 

A. -Sax. bAr (boor), one of the lowest class of 

freemen, a husbandman ; also a bower. 
O. Nor. bczr. byr {Y>d^r\.-l^ot\\. and Swed. by), 

settlement, farmstead, village. 

A. -Sax, carl (Scand. karl), ceorl (churl), one of 
! the lowest class of freemen, a husi)andman. (A 
I slave was a theo-o or a thnil, whence the 
surnames Thew and Thrale). 

Lat. castra (pi. of caslruin), camp ; whence 
A. -Sax. ceaster, city. 

(i) A-Sax. dcBl, Scand. dal, dale; (2) A.-Sax. 

dtel, Scand. del, deal, allotment, 
(i) A.-Sax. ig (ly), O. Nor. ey, island, low 

riparian tract ; (2) A.-Sax. ed, O. Nor. a, river. 
A.-Sax. gcers, grass, 
(l) A.-Sax. geat, gate, passage, road ; (2) A.-Sax. 

gdt, goat. 

(i) A.-Sax. qraf, gxo\& ; (2) A.-Sax. grce/, O. Nor. 
grof (Dan. grav, Swed. graf), trench, ditch. 

(i) A.-Sax. heath, O. Nor. hall{r), slope, hill, 

corner ; (2) A.-Sax. heatl, hall, 
(i) A.-Sax. hdin, home; (2) A.-Sax. ham{m), 

piece of land, often hemmed in by the bend of 

a river. 





















Trop / 

(i) A. -Sax. kara, Scand. /lare, hare (combined 
with grove, wood, field, ley, etc.); (2) A.-Sax. Adr, 
O. Nor. Aarr, grey ; (3) O. Nor. /lar, high ; 
(4)? A. Sax. kere, O. Nor. /lerr, military. 

A.-Sax. heorde, herd ; occasionally A. Sax. per- 
sonal name Heard = brave. 

A.-Sax. hege, haga, hedge, enclosure. 

A.-Sax. ho, hSh, hough, heel 
formed like a heel. 

point of land 
Nor. holm, river-island, low 

A.-Sax, and O. 
riparian land. 

A.-Sax. holt, copse, wood. 

A.-Sax. hyrst, copse, wood. 

A.-Sax. suffix denoting 'son of,' in pi. 'descend- 
ants of.' 

O. Nor. kirkja (Dan. -Norw. kirke, Swed. kyrka), 
A.-Sax. circe, church. 

A.-Sax. ledh, meadow, pasture. 

A. -Sax. hl(£W, (burial) mound, hill. 
O. Nor, niel[r), sandhill, sandbank. 
O. Nor. ties (Dan. -Norw. ties, Swed. tids), 

A.-Sax. ticess, headland, promontory. 
A.-Sax. tieother, lower, 
(i) A.-Sax. ofer, upper ; (2) A.-Sax. 6fer, bank, 


A.-Sax. p6l, O. Nor. poll{r), pool. 

O.-Nor. sk6gr (Swed. skog, Dan. -Norw. skov), 
A.-Sax. sceaga, 'grove,' 'wood.' 

A.-Sax. sceot, scedf, angle or corner (of land), field ; 
sometimes corruption of A.-Sax. s-holt, the s 
being the genitive or possessive suffix of the first 
element of the name ; holt = wood. 

A.-Sax. Stan, stone, rock ; castle. Bosworth and 
Toller {A.-Sax. Dict.)have collected the follow- 
ing significations of stdn : i. stone as a material, 
ii. a stone, iia. a stone for building, iib. a stone, 
natural or wrought, serving as a mark, iic. an 
image of stone, iid. a stone to which worship 
is paid, iie. a stone containing metal, iif. a 
precious stone, iig. stone (med.), iii. a rock. 

O. Nor. sta(h{r), A.-Sax. stede, place. 

A.-Sax. siith, south. 

O. Nor. thittg, council, parliament. 

O. Nor. thorp, farm, hamlet, village. 




Wal (prefix) 

Wall (suffix) 


O. Nor. tkveit, piece of land, clearing. 

A. -Sax. tiSn, enclosure, farmstead, manor, village ; 
mod. town. (Also found in Icelandic.) 

(i) A.-Sax w^fl//, wall; (2) A. -Sax. twa/^/, forest, 

O. Nor. vollr, field. 

Most commonly prob. A. -Sax. 7ver {O. Nor. ver 
or vorr), weir, dam ; fishing-place ; (?) landing- 

(i) A. -Sax. zvic, habitation, station, creek, bay; 
(2) O. Nor. vik, creek, bay. 

A. -Sax. weorihig, farmstead, estate. 



Herleshala \ 
Heleshale j 


(West) Derby. 1 

Ravensmeols. [pool. 

Smithdown (Lane), Liver- 


Ince (Blundell) 


(Up) Holland. 

(Down) Holland. 



^ Domesday records that there belonged to the manor of Derbei six 
berewicks, or subordinate manors, which it does not specify. These 
are presumed to have been Litherpool or Liverpool, Everton, Thing- 
wall, Garston, part of Wavertree, and Great Crosby. 



Domesday. Modern. 

Liderlant ) 
Literland ) 





(North) Meols. 








\ Orr^ll 

Otringemele j --— 





Schelmeresdele Skelmersdale. 













Walton (on-the-Hill) 






Domesday. Modern. 








































Ness, or Nesse. 






(Over) Pool 


Poulton (Lancelyn). 








Saughall, Soughall. 












Thornton (Hough). 






A.-S., A. -Sax. 

Anglo-Saxon or Old English. 



Modern Danish. 




Du., Dut. 



E. Eng. 


Early English. 












German, i.e., New High German. 













M.E., Mdle. Eng. 


Middle English (i2th to 15th cent.) 

Mod. Eng. 


Modern English. 

Nor. Eng. 


Northern English. 

O. Eng. 


Old English. 

O. Fr. 


Old French. 

0. Fris. 


Old Frisian. 

O. H. Ger. 


Old High German. 

O. Nor. 


Old Norse or Icelandic. 

O. Sax. 


Old Saxon. 



Scandinavian, ?.£., common to t 
Scandinavian languages. 











A. -Sax. and O. Nor. S, i> = th. 


With the possible exception of London, the name of no 
EngHsh town has excited so much discussion, and been the 
cause of so much philological brain-cudgelling, as has that of 
Liverpool. As early as the latter half of last century, the 
magazines began to take up the tangled etymological thread 
which had been but little more than touched in passing by 
the chroniclers of the i6th and 17th centuries. At a later 
period the columns of Notes and Queries were from time to 
time, and are still occasionally, opened to expressions of 
opmion upon this evergreen question of the origin of the 
first two syllables of ' Liverpool.' 

Let us glance at the earliest recorded spellings of the 
name. The most ancient deed in which it is found belongs 
to the time of Richard I (11 89-1 199): the form here is 
Leverpol. In King John's charter, 1207, we have Liver- 
pul ;^ in that of Henry III., 1229, Levereptil ; but in the 
Testa de Nevill (fol. 371), in a part which bears distinct 
evidence of having been written in the reign of King John, 
we find the form Litherpol. An analysis which I have made 
of the spellings of the name of the city in 36 of the earliest 
(13th cent.) Moore charters and deeds relating to Liver- 
pool- gives the following result Leverpol 1, Liverpiil i, 
Liverpo/ 2, Liverpool 8, while Lyverpol occurs no fewer 

' Livcrpiil K also the spelling in the Pipe Roll of 10 John (1209), 
membrane 10. 

- The Moore Charters and Documents Relating to Liverpool : 
Report to the Finance and Estate Committee of the City Council ; 
Part I., by Sir J. A. Picton, 1889. I am indebted for a copy of this 
first part of the Report (the only part so far printed) to Mr. T. N. 
Morton, Clerk of the Records, Liverpool. 


than 2)2) times. In the State records of the early part of 
the succeeding century the speUings are almost uniformly 
with V. Thus in the Close Rolls we have the following 
forms : 1314, Lyverpol ; 1323 and 1328, Liverpol. In the 
Open Rolls: 1330, Liverpool (as to-day); 1333, Liver- 
pull dind Liverpole ; xTyTi"], Leverpol. In Rymer's Foedera : 
1323, Liverpol; 1327, Lyverpol ; 1336, Liverpull. And 
the V form continued to be by far the more usual until the 
name definitively settled down into its present spelling ; 
among the most notable exceptions to the v rule (apart from 
the instance already mentioned) being the Letherpole of the 
Ministers' Accounts, Duchy of Lancaster, 1509 •} the Lither- 
pole of the Calendar to Pleadings, Duchy of Lancaster, i. 
183 (1547) ; the Litherpole and Z///^^^C(?/6' of Camden ;- the 
Litherpoole of a miscellaneous record of the Duchy of 
Lancaster, dated 1640-41 f and the Litherpool oi Baxter. ■* 

For Camden's "in Saxon Liferpole" there is not the 
slightest authority, the earliest existing document in which 
the name occurs belonging, as we have seen, to the end of 
the 1 2th century, although there is little doubt that Lither- 
pool or Liverpool was one of the six unspecified berewicks 
mentioned in Domesday as being attached to the manor of 

^ Selby's Lane, and Chesh. Records Preserved in the Public Record 
Office, London (Lane, and Chesh. Record Soc), 1882-83, i. 100. 

- In the first edition of Camden's Britannia (1586, p. 429) the 
passage relative to Liverpool ran : ". . . ubi Litherpole floret, vulg6 
Lirpole, a diffusa paludis in modum aqua, ut opinio est, nominatus, 
qui commodissimus et usitatissimus est in Hyberniam traiectus, 
elegatia et frequentia, quam antiquitate celebrior." In the edition of 
1607 (p. 612) this paragraph is modified as follows : " . . . ubi Lither- 
poole patet, .Saxonice Liferpole, vulgo Lirpoole," etc. Gibson's trans- 
lation (17/2, ii. 146) runs "... Liverpool, in Saxon Liferpole, 
commonly Lirpool ; so called (as it is thought) from the water spread 
there like a fen. It is the most convenient and usual place for setting 
sail into Ireland, but not so eminent for antiquity as for neatness and 
populousness." Gough translated it (1789, iii. 12S) : "... Lither- 
poole, Saxon Liferpole, commonly called Lirpoole, from a water 
extended like a pool, according to the common opinion, where is the 
most convenient and most frequented passage to Ireland ; a town more 
famous for its beauty and populousness than its antiquity." 

^ Selby's Lane, and Chesh. Records, etc., i. 33. 

■* Glossarium Antiquitatum Britannicarum, 2nd ed., 1733, p. 213 : — 
"... hodiernum vero loco nomen Lither-pool est, sive Pigra palus'' 
(The present name is really derived from the situation — Litherpool or 
' sluggish water '). 


West Derby (Derbei). Of the other i6th century chroniclers 
Harrison has :^ " Lirepook, or as it was called of old, 
Liverpoole haven " ; and Leland -?■ " Lyrpole alias Lyver- 

Notwithstanding the much greater frequency of the 
spelling with v compared with that with th (the forms 
without either are, of course, merely slurred renderings of 
the proper name), it is impossible to say with certainty, from 
the available historical evidence, whether the original form 
of the name of the city had liver or lever,^ lither or lether ; 
but the contiguity of Litherland would almost seem to 
indicate that the spelling with /// was the primitive one. 

In an article in the Supplement to the Lady's Magazine 
for 1774 (p. 676), it is asserted that "the right spelling" of 
the name of the Mersey port " is ' Leverpool ' " ; but the 
article is simply based on Enfield's ' I>everpool,'^ which was 
published in that year. Enfield offers no definite etymology 
of the name ; he merely refers (chap. I.) to the hypothetical 
derivations from the fabulous liver bird, or the seaweed 
liver, or the Lever family, without manifesting a preference 
for any one of them. In the Gentlemari s Magazine, 
vol. Ixxxvii. (1817), pt. ii., p. 508, Mr. W. R. Whatton, of 
Manchester, traces the etymology to a conjectural A.-Sax. 
Lifiepul, which he translates as ' still or quiet lake ' ; and 
the elder Baines'' seemed inclined to agree with his coadjutor 
hereon. But the A.-Sax. /?«<' is the Mod. Eng. ' lithe,' which 
would give ' Lithe-pool,' instead of Litherpool. 

Enfield's successor, Troughton, experienced not the least 
difficulty in fixing the etymology of ' Liverpool ' — or ' Lither- 
pool,' as he preferred to spell the name. "The word iither,^' 
he says,*^ " signifies lower. Litherpool means Lower Pool. 
Hence the name of the village Litherland, or ' Lower Land ' ; 
and of a passage, yet called Litherland Alley, in the neigh- 

1 Holinshed's Chronicles, Hooker's ed., 1587, i. 84^. 

* Itinerary, Ilearne's 2nd ed., 1744, vii. 44. 

^ The student should not overlook the correspondence entitled 
" Leverpool or Liverpool ?" which was initiated in the Liverpool Courier 
on June 7, 1889, by Mr. Ellis Lever. 

'' An Essay towards the History of Leverpool, based on Mr. George 
Perry's papers, by William Enfield, 1774. 

^ Hist, of Lancashire, 1836, iiii. 55. 

^ Hist, of Liverpool, 18 10, p. 20. 


bourhood of Pool Lane." The A. -Sax. word for ' lower,' 
however, is neothor, ' nether.' 

The younger Baines worked on somewhat different lin- 
guistic lines. He was inclined to think that the Z/der and 
Zt'fer of Domesday,^ the Ze7'er of the reign of Richard T., 
the Lt'f/ier of Testa de Nevill, etc., were " all originally the 
same word, and that they are derived, as has been suggested, 
from the old Gothic word h'de or /if/ie, the sea, or from 
some of the words formed from it : as /id and /ifer, ' a ship '; 
////le, 'a fleet of ships'; lithesnian, *a seaman.'"- But, as 
Professor Skeat points out, Gothic has no such word as 
liiSe or lithe, and it does not mean ' the sea ' in Anglo- 
Saxon. It is an adjective, and signifies * gentle ' — Mod. 
Eng. lithe, 'pliant.' Lid, ' a ship,' is from a different root, 
and has nothing to do with it. 

The latest historian of Liverpool, Sir J. A. Picton, found 
the question altogether too knotty for solution, although he 
possessed infinitely greater philological knowledge than 
any of the preceding historians of Liverpool. " The name 
of Liverpool," he observes, " is even more enigmatical than 
the seal, and has hitherto baffled all investigators in 
endeavouring satisfactorily to account for its origin. That 
the name was originally applied to the water rather than to 
the land appears to be agreed upon all hands. The 
embouchure of the small stream was called the Pool down 
to the time of the formation of the Old Dock."^ Sir James 
afterwards pointed out that the notion of giving the name 
liver to a bird (which constitutes the popular etymology) was 
quite unauthorized, the symbolic Liverpool bird being 
originally the Eagle of St. John. On this point Dr. Colling- 
wood wrote, over thirty years ago : " In both the ancient 
and the modern seal [of Liverpool] we have a bird which 
has neither the long legs of a heron nor the long neck of a 
liver (?), but is as good a representation of a dove bearing 
an olive branch as we could expect to see in such a situa- 

^ This refers to Litherland, there being (as already stated) no men- 
tion of Liverpool in Domesday, except that the Esmedune of that 
Survey is assumed to have been the name of a hamlet situate in or near 
the present Smithdown Lane, Liverpool. 

- Hist, of Liverpool, 1852, p. 58. 

' Memorials of Liverpool, 1875, i- 40* 


tion. The etymology of the name of Leverpool, or Liver- 
pool, is doubtless topographical rather than heraldic or 

Now let us note the most recent expert opinion upon the 
question of the etymology of ' Liverpool' 

In A^ofes and Queries, February 29, 1896, p. 173, Pro- 
fessor Skeat writes : "... we see that the name /h'er was 
certainly applied to some kinds of the iris and the bulrush 
which grew in pools, whence it appears that liver-pool meant 
originally neither more nor less than ' a pool in which 
livers grew,' meaning by h'ver some kind of waterflag or 

To this Canon Taylor replied {Notes and Queries, March 
21, 1896, p. 233): "Professor Skeat concludes that the 
liver in Liverpool, Livermere, and Liversedge denotes some 
kind of iris, waterflag, or bulrush, which grew in pools or 
meres. An obvious difficulty is that while the mere at 
Livermere was a freshwater mere in which waterflags or 
bulrushes might grow, the pool at Liverpool is [Canon 
Taylor was evidently unaware that the pool is filled up] a 
saltwater pool, in which no such growths are possible." 
The Canon adds that the meaning of ' Litherland ' would 
throw light on the signification of ' Litherpool ' or ' Liver- 

But Professor Skeat had already (March 4) written to me 
as follows : "I am aware that some documents have 
Lither . . . ; but if that is right the name also makes 
perfect sense as it stands, being the O. Eng. lither, an 
extremely common word, meaning ' bad,' and so ' dirty ' or 
'disagreeable.' Shakespeare uses it in the sense of 'stagnant' 
or ' sluggish,' as applied to the air, in a passage explained 
by me some years ago " (i Hen. VL 4, 7, 21). And again 
(March 9) he wrote : " It \litJier or any of its dialectic 
variations] is used all sorts of ways. Thus lether sti is ' a 
bad road.' I see no great difficulty over lither-pooiy 

We have already observed that Baxter explained Lither- 
pool as Pigra palus, or 'sluggish water.' Lither is, more- 

^ The Historical Fauna of Lancashire and Cheshire, Proceedings 
Lit. and Phil. Soc. Liverpool, vol. xviii. (1863-64), p. 170. 


over, well known as a Northern dialect-word, meaning 
' sluggish,' ' idle,' ' lazy.'^ 

If, however, we were to accept Professor Skeat's theory 
of a ' dirty pool,' what are we to say of ' Litherland ' 
(Domesday Liderlant and Literland) ? I can scarcely think 
that it was ever intended to designate this particular district 
' bad land.' We must, too, pass by the would-be derivation 
from a hypothetical Celtic Llyrptull (Welsh llyr, ' brink,' 
'shore,' 'sea'). The A.-Sax. piil (O. Nor. pollr), 'pool,' 
is admittedly from the Cymric ptvU^ as a general thing ; but 
it clearly does not follow that Liverpool is necessarily of 
Welsh origin. Besides, pwll in Welsh bi-elemental place- 
names, in accordance with the genius of the language, is 
invariably a prefix, as in ' Pwllheli '; and, furthermore, while 
(as we have observed) ' Liverpool ' or ' Litherpool ' can be 
slurred into Lirpool, it is hardly conceivable that Llyrpwll 
could be extended into ' Liverpool ' or ' Litherpool.' 

It would seem that we must rather look to the Teutonic 
lilith, ' a slope,' to supply the much-sought-for etymology. 
This word is cognate with Lat. clivus^ ' a gentle ascent,' 
which exactly tallies with the physiography of Litherland, 
and the course of the old Pool. It is, however, somewhat 
doubtful whether the A.-Sax. hlith or the O. Nor. hlWi was 
the original etymon. Were we to accept the former, it 
would be necessary to assume that the er constituting the 
second syllable of ' Litherpool ' and ' Litherland ' was a 
phonetic intrusion, owing to the difficult vocal combination 
presented by thp. This is, perhaps, a rather unusual assump- 
tion with respect to a place-name ; but we are not altogether 
without historical evidence for such reasoning. Kemble 
says,'"^ referring to the common Anglo-Saxon legal terms, 
and those relating to custom^and usages occurring in 
charters : " The Normans could not even spell the words, 
hence they write griderbryce, friderbryce, ior grihbryce [breach 
of covenant], fri^bryce [breach of peace], their own contrac- 
tion for der bearing a distant resemblance to the Saxon S 

' Lither is glossed 'idle,' 'lazy,' in both the Lancashire Glossary 
(1875-S2) and the Cheshire Glossary (18S4-S6) of the English Dialect 
Society, as well as in Sir J. A. Picton's South Lancashire Dialect, 
Proceedings Lit. and Phil. Soc. Liverpool, vol. xix. (1864-65), p. 35. 

* Codex Diplumaticus, i. xliii. 


\dhox //;]."! ^This is ^'exactly what may have? happened 
in the case of Litherland, which in Domesday is spelled 
Liderlant and Literland^ and later, Litterland^ owing to the 
difficulty experienced by the Norman French in the articu- 
lation of the th or dh sound — th as in ' thank,' dh as in 
' then.' 

On the other hand, if we accept the Norse hlitJi^ which 
figures in various Scandinavian place-names, there is no 
necessity to assume a phonetic intrusion, because the genit. 
sing, case of the word in that language is hlithar, compared 
with hlithes in Anglo-Saxon. In Old Norse we find hlithar- 
brun, ' the edge of a slope,' hlithar-fotr, ' the foot of a slope,' 
so that there would not be much difficulty about hlithar- 
la}id, ' the land of the slope ' or ' the slope land,' and 
hlithar-pollr^ ' the pool of the slope ' — in fact, Hliiharldtid 
{lofid, pi. of land) occurs in the Icelandic record, the 
Landnamabok.- It is scarcely necessary to point out that 
the Norse element in the vicinity of Liverpool was very 
strong ; and the appropriation of creeks and sea-pools, and 
of land adjacent to the coast, was a well-known character- 
istic of the ancient Norwegians and Danes, as, indeed, the 
name Vik-\ng implies. It should be mentioned that this 
derivation from the Old Norse is that which recommends 
itself both to the Lecturer in Icelandic in University 
College, Liverpool (Rev. J. Sephton), and to the Professor 
of Latin in that institution (Dr. Strong). 

It only remains to account for the presumed transition 
from the th to the v sound in the case of Liverpool. This 
is a very simple matter. Max Miiller published'^ a diagram 
showing how easily ih was changed into / or v, and quoted 
a well-known German authority^ in reference hereto. Dr. 

^ As to intrusive r see Skeat's Notes on English Etymology, s.v. 
listre, Transactions Philological Soc, 1885-86, pp. I sqq. ; also a 
paper by Dr. Stock on the Influence of Analogy as Explaining Certain 
Examples of Unoriginal / and r ; ibid. pp. 260 sqq. ; and compare 
Nidderdale, i.e., the dale of the River Nidd. 

2 Compare also Lithar-fylki {fylki =folk), mentioned in the 12th 
cent. Danish History of Saxo Grammaticus, lib. v., which place is 
identified by Holder (Saxonis Grammatici Gesta Danorum, ed. 1886, 
p. 702) with Lier, near Drammen, in Norway. 

■* Science of Language, 2nd ser., 1864, p. 175. 

^ Arendt, Beitrage zur Vergleichenden Sprachforschung, i. 425. 


Morris wrote •} "The spirants interchange with one another : 
f=th. Children often S3.y fu/nd for ' thumb.' Cf. 'dwarf,' 
Mdle. Eng. dwerth and dwerg=0. Eng. tkiueorh ; Russian 
Fedor = Theodore." With the uneducated Londoners, too, 
th=f ox V ('with' is pronounced wiv), and this is attri- 
butable to Norman-French influence. We have also such 
instances as that supplied in ' The Carved Stones of Islay,' 
where Mr. R. C. Graham refers to a slab with the inscrip- 
tion, "John Heves, marchand in lever [leather], 1702.'' 
The cognate form in Latin {clivus) of Teutonic hlith has 
already been noted 

The reason why ' Litherland ' did not suffer the mutation 
with ' Litherpool ' is, I think, sufficiently obvious. Lither- 
land, until comparatively recent times, has been a remote 
and secluded village. Liverpool, as a seaport, after King 
John developed an interest in it, became more and more 
subject to extraneous and distracting influences. On the 
same principle the Icelanders have, by their isolation, pre- 
served the language of their Norse ancestors almost intact, 
while their brethren, the inhabitants of easily-accessible 
Scandinavia, have broken it up into modern Norwegian, 
Danish, and Swedish. 

^ Historical English Grammar, 1875, p. 44. 



Abram. — This name affords a curious instance of what 
some philologists term 'folk assimilation.' The ancient 
designation was Edbiirgham or Adbi/ri^ha/n} The place 
was therefore originally the ham or home of a Saxon pro- 
prietress named Eadburh {durh being a termination of 
feminine names only). 

Aigburth. — The name of the headquarters of the 
Liverpool Cricket Club represents ' the place of the oaks ' — 
A. -Sax. dc, 'oak,' combined with a derivative of beran, 
'to bear,' 'to produce.' In the Ministers' Accounts of 
Whalley Monastery, 28-29 Henry VHI. (1537-1538), Lane, 
and Chesh. Rec'd Soc, vol. vii., 1S82, we have the spelling 

Ainsdale. — The Domesday Ei?i2dv€sdel^ and the sub- 
sequent variations of that name, all imply ' the dale-land 
(A.-Sax. did) belonging to Einulph or Eanulf,' which was 
a fairly common Saxon name, contracted from Earnwulf, 
that is, Eagle-Wolf. 

Aintree. — The name of this famous racing resort, 
the Rev. Ed. Powell, of Lydiate — who has devoted much 
time to the study of old deeds relating to the Orms- 
kirk district — informs me, is nearly always spelled Aintree 

^ Chancery Rolls, 7 Henry IV. (1406), membrane 6 ; Lane, and 
Chesh Rec'd Soc, vol. viii., 1883. 



or Ayntree^ in ancient documents ; and as to this day there 
is a conspicuous absence of large plant-hfe in the district 
there seems no reason to doubt that Aintree (A.-Sax. 
dH'trebiv) means simply ' One Tree.' In fact, upon this 
very point Professor Skeat writes to me : " We have ' Ain- 
sty,' single or one path (narrow path for one) ; and I know 
a ' One-Tree Hill ' myself." 

AUerton. — This is a common English place-name, 
meaning ' the alder enclosure or farmstead ' — A.-Sax. 
air, ' alder,' combined with the usual ti'in. The Allerton of 
the West Derby Hundred occurs in Domesday as Alretiine. 

Alt.— The name of this stream has a certain Teutonic 
appearance, alt being modern German for 'old' — A.-Sax. 
eald?' Alt is the name of a river in Central Europe, and 
it has an alternative appellation, Aluta, which might recall 
the A.-Sax. alutan, ' to bend,' or aledt, ' bent down,' with 
reference to the peculiar and decided way in which the 
Lancashire Alt bends downwards or southwards where it 
approaches its embouchure. But there can be no doubt 
that, as in the case of so many river-names in England, 
' Alt ' must be referred, like ' Douglas ' (black water), the 
name of a neighbouring river, and ' Gowy ' (Welsh gwy, or 
wy, ' water '), a river at the southern end of the Wirral pen- 
insula, as well as 'Dee,' to Celtic sources — in this instance 
to the Gael, alt, ' a stream ' (altd?i, ' a streamlet '), a name 
which, naturally generally combined with an adjective-suffix, 
occurs frequently in Gaelic-speaking districts. 

Altcar, near Formby, is enclosed on three sides by 
the waters of the Alt and its affluents ; and car is a common 
word in this part of Lancashire for ' moss-land ' — Mdle. Eng. 
car or ker, O. Nor. ktarr, SBwed. kiirr, ' marshy ground.' 
Baincs ('Lane.,' 1870, ii. 405) says: "Altcar seems to be 
the Domesday Acrcr, half a carucate of waste land held by 
Uctred ; but no other mention is made of this place until 
21 Edward L (1293), when an action was tried between the 
King and the Abbot of Mira Vallis, or Merivale, as to the 
right of the latter to one carucate of land in Altekar . . . 

1 Ayntre occurs in the Nonarum Inquisitiones {temp. Edw. III.) 
fol. 40^ 

- The Scandinavian lan£juap;es differ from the Germanic in retain- 
ing a different root for ' old ' — O. Nor. gaiiiall. 


As appears from several maps, a hamlet called Altmouth 
formerly existed at the mouth of the River Alt ; and it is 
supposed to have been overwhelmed by inroads of sand, 
similar to those which drove the inhabitants of Formby to 
remove their church and village inland. 'l"he remains of 
houses have been found buried in the sand immediately 
contiguous to the railway station, Hightown, which is on 
the site assigned to Altmouth in the township of Ince." 

Anfleld. — A part of Everton, originally Plangfield, or 
Hanging-fields, in allusion to the deeply sloping or hanging 
nature of the ground. Syers says :^ " In the 3rd Henry VH. 
anno 1488, an inquisition was taken at Walton, which shows 
that the boundary of the south part of Walton, ' beginning 
at Carton Cross, and following to Darling Dale, and to the 
cast end thereof, and so over the Breck, by an ancient 
ditch on the lands of Everton, called Ilangfield, on the south 
part of the common of pasture of Walton,' " etc. He adds : 
" All the lands of Everton were known by the names of 
Hangfield, Whitefield, and Netherfield "; and he then 
appends a footnote : " This word [Hangfield] is frequently 
written ' Hongfield,' and by some writers ' Honghfield.' 
I prefer Hangfield, that name being derived from ' hanging, 
or sloping field.' To strengthen the propriety of my ortho- 
graphy in this particular instance, it may be as well to state 
that in Gore's paper of 26th July, 1810, certain fields of 
Walton are advertised as follows : ' Fields in ^Valton-on- 
the-Hill, called Hanging-fields.' " 

Arbury. — The Testa de Nevill has Herbury (fol. 396) 
and Herelmr (fol. 398''), which is exactly how a Norman 
scribe would write down the spoken ' Arbury ' ; so we must 
consider the h in the Testa to be intrusive or unoriginal. 
It is difficult, without evidence of Saxon spelling, to give 
the meaning of Ar with anything approaching certainty. 
I do not favour A. -Sax. (crra., 'old,' but rather think that 
Arbury was perhaps originally a simple earth-fortification — 
A. -Sax. eorth-lmrh {cf. A.-Sax. //twV// = Mod. Eng. 'hearth'); 
and this derivation would seem to be confirmed by the form 
Erdluiry, which is found in a Duchy of Lancaster document 
of 18 Henry VII. (1502-3). 

^ Hist, of Everton, 1830, p. 19. 


Atherton. — This place occurs in the Testa de Nevill 
(fols. 396, sgS*", and 408) as Aderfo?i. d and 'S (th), it 
is well known, were constantly interchangeable in Early 
English : the word ' gather ' is from the A.-Sax. gaderia?i, 
while ' murder ' is from A.-Sax. mortJior. A small stream, 
which at one point expands into a pool, runs through the 
township, and there seems no reason to doubt that Atherton 
is an equivalent of Broughton or Brocton, the iihi or farm- 
stead by the brook, and is derived, as to its first element, 
from the A.-Sax. ddre (compare Swed. dder), ' fountain,' 
'watercourse.' It is true that Bosworth ('A.-Sax. Diet.,' 
1838), on the authority of Lye(' Diet. Sax., etc.,' Manning's 
ed., 1772), gave isf/ier as meaning *a field,' which would 
make Atherton a synonym of Felton, the ///« on an open 
field or plain ; but Lye took txf/ier from ^.Ifric's 'Colloquy,' 
where it is glossed ager (field), while later vocabularists 
declare ather, as it there occurs, to be a mistake for oicer, 
* field.' Thus Wiilcker, editing Wright,^ prints cecer in the 
' Colloquy,' and adds a note : " The MS. reads distinctly 
cether, which is no doubt an error for cEcer." 

Aughton. — Aughton, or Aghioft, as the name was 
spelled in the reign of Edward II. (1307-1327), might at 
first sight appear to be referable to the Gael, auch and the 
Ir. ag/i, augh {ac/tad/i), ' field '; but seeing that the spelling 
Acton occurs in the times of John and Henry III. (1199- 
1272), and bearing in mind that the ch in the Domesday 
Achctun (as our Aughton figures in the Norman Survey) 
generally represents hard c (k), it seems but reasonable to 
assume that Aughton was simply ' the oak farmstead,'^ 
A.-Sax. dc, ' oak' — ^just as the Aughton in Yorkshire was. 

Bamfurlong, the name of a village near Wigan, is 
literally ' the tree furrow-length' — A.-Sax. beam {Ger. baum), 
'tree'; furh, ' furrow '; /a?ig, ' long.' A furlang, or furlong, 
appears to have originally been a square as well as a long 
measure, "having," says Kemble," "determinate length and 
width, and forming a fixed portion of an acre." 

^ Anglo-Saxon and Old English Vocabularies, 2nd ed., 1884, p. 90. 

- It should, however, be mentioned that the spelling Asshton occurs 
in the Survey of 1320-1346, Chetham Soc, vol. Ixxiv., 1868. The 
ash was the favourite tree of the Anglo-Saxons. 

^ Codex Diplomaticus, iii. xxv. iSut see footnote p. 45. 


Bankhall. — Bank Hall, close to the l/ank of the 
Mersey, was long the residence of the ancient Liverpool 
family of More or Moore. No trace of the mansion now 

Barton. — Domesday spelling Bartune. Barton is a 
very common English place-name, derived from the A. -Sax. 
bere-tun, ' barley yard,' ' grange. ' It is distinct from Burton, 
which see in the Wirral Section. 

Bedford. — The name of this township does not occur 
in the earliest extant records ; it is not mentioned in 
Domesday or the Testa de Nevill. It seems, however, to 
have formerly been spelt Bcdefoni ; and the original appel- 
lation may have been identical with that of the like-named 
Midland county town, which meant ' the ford of Bedica,' 
the sequence in Anglo-Saxon deeds relating to this impor- 
tant place being briefly : Bcdicatiford {Bedican, genit. sing. 
of the personal name Bedica), Bedanford, Bedeford. Or the 
Lancashire Bedford may originally have simply been Bedan- 
ford, ' the ford of Beda.' 

BickerstafiPe. — The earliest recorded spelling of this 
name, Bikersiat, seems to be in the Testa de Nevill (fol. 
402*^'). In ancient Burscough MSB. Bykyrstath repeatedly 
occurs, and even as late as 1517 we find Bekerstath, 
although the termination staff ox staffe had apparently been 
used before that date. The first portion of the name 
is probably from a personal appellation, while the last 
syllable must be attributable to the O. Nor. stathr (A. -Sax. 
stede) ' a place.' As a personal name Biker or Bicker may 
be from two sources. The O. Nor. Inkarr, ' a large drink- 
ing cup,' is still with us in the Scot, bicker and the Eng. 
beaker ; and Jamieson^ says that bikarr " was the term used 
to denote the cup drunk by the ancient Scandinavians in 
honour of their deceased heroes." But more likely the 
origin of the personal appellation is to be found in the 
O. Nor. byggia (Dan. bygge, Swed. bygga), ' to build,' whence 
the Scot, and Nor. Eng. big or byg, ' to build,' and biggar, 
' builder.' 

^ Scottish Dictionary, under bicker, quoting Keysler, Antiq. Septent., 
pp. 352-354- 


Billinge. — This was a settlement of the Saxon BilHng 
family,^ and, according to the canon now accepted, an 
original, or parent, or ancestral settlement, as distinct from 
the offshoot which would be imphed by a Billington or a 

Birkdale. — On the face of it this name would simply 
represent 'the dale of the birches' — O. Nor. l>idrk, 'birch,' 
and Scand. dal; but as the earliest found spelling is 
£reck-en-le-Dak, which occurs in a deed of 6 Edward I. 
(1278)— Duchy Records— by which Sir Robt. Blundell, son 
of Adam, lord of Aynosdale (Ainsdale), released certain 
lands to his son, birches have apparently nothing to do 
with the etymology of Birkdale. The exact meaning of the 
dialect-word ' breck ' is somewhat doubtful ; it seems to 
vary in different parts of the Anglian country to which it 
is confined. Thus the ' Hist. Eng. Diet.' notes from 
local authorities that in Norfolk and Suffolk ' breck ' 
signifies a large field, while in Northumberland, etc., it 
represents 'a portion of a field cultivated by itself.' Some 
believe that a breck was, in all Teutonic countries, originally 
a piece of land that was either temporarily or permanently 
idle or waste in the midst of a cultivated tract. The Mod. 
Eng. equivalent apjjears to be 'brake' or 'bracken,' which 
are connected with the Dan.-Norw. bregne, 'brake,' 'fern'; 
Dan.-Norw. brak, 'fallow'; Swed. brdken, 'brake,' 'fern'; 
Ger, brack, ' fallow,' etc. We should not, however, overlook 
the Norse word brekka, 'a slope,' which is frequently found 
in Icelandic local names and will doubtless account for some 
English ' Brecks.' 

Blundellsands. — This place, topographically, is prac- 
tically identical with Crosby, where is the residence of 
the Blundell family. " Within the last forty years," say 
Baines's editors,^ " Great Crosby has very considerably 
increased, having become a regular place of residence for 
Liverpool merchants ; and the present proprietor of Little 
Crosby has appropriated a long track of sandhills on the 
coast for building purposes, under the name of Blundell- 

^ Birch's ed. (1876) of Kemble's Saxons in England, i. 458. 
^ Lane, 1870, ii. 398. 


Bold, Bootle. — The names of these places present 
no etymological difficulty. The A.-Sax. l>o/d, bolt^ botl, 
meant 'house,' 'dwelling,' 'village.' In Domesday the 
name of the place which is now called Bootle was spelled 
Boltelai. Upon this point Prof. Skeat writes to me : " I 
suppose that, in Boltelai, the boltel alone represents 
' Bootle,' and the ai is a suffix." In the Testa de Nevill, 
fol. 403 (for Bootle), we have the spelling Both; in a 
document dated 1541 B otill occvixs. Leo's identification of 
A.-Sax. boil and bold with Ger. battel} is now shown not 
to be correct. 

Bryn. — This Lancashire place-name has no connection 
with the Wei. bryn, 'hill.' It is probably of Norman-French 
origin, as Alan le Brun (O. Fr. brun, ' brown '), who is men- 
tioned in the Testa de Nevill (fol. 403) as holding by 
ancient tenure two bovates of land for 6s. of Henry de Lee, 
Sheriff of Lancashire in 1274 and 1282, is reputed to have 
given his name to Brun or Bryn Hall. The ultimate source 
of Fr. brun is, however, the O. H. Ger. brun, so that the 
origin is Teutonic. See Stappers's ' Diet. Synopt. d'Etymo- 
logie Fran^aise,' 1894, p. 506, and compare A.-Sax. brun, 
Ger, braim, ' brown.' 

Burscough. — The spelling in 1 28 $ yf^s Bi/rscho7a ; in 
the Survey of 1320-1346, Chetham Soc, vol. Ixxiv., 1868, 
we have Burschoge. The latter element in the name is 
allied to the Mdle. Eng. shaive, A.-Sax. sceaga, ' thicket,' 
' small wood,' and is from the O. Nor. skbgr (Dan. skov, 
Swed. skog). The first element may be from a personal 
name, or it is perhaps connected with Mod. Eng. ' bur ' or 
' burr'; cf. Dan. borre, 'burr,' 'burdock.' Or the grove may 
have contained a bur, or bower. 

Burtonwood. — Burton = A.-Sax. bur-tiin, ' the boor 
farm ' ; compare Chorlton. But see the Wirral Burton. 

Ohildwall. — The Domesday spelling is Cildeuuelle, 
which shows that the name of this township was pronounced 
at the end of the eleventh century practically as it is to-day ; 
in 1259 Childwalle occurs ; in the Testa de Nevill we have 
Childeivale and Childeivall. Other renderings are Chelde- 

1 Die Angelsachsischen Ortsnamen (Rectitudines Singularum 
Personarum), Halle, 1S42, p. 36. 


zvell, Chidewell, Chidwall {i62:\), Chilwell {i']6o), and even 
Kydeivell. The second syllable is homonymous with the 
-walloi 'Thingwair — O. Nor. w//r(dat. sing, velli), 'field'; 
and we have the O. Nor. kelda (Dan. kilde, Swed. kdlla, 
Nor. Eng. keld), ' well,' ' spring.' The Mod. Eng. equi- 
valent of Childwall is therefore ' Wellfield ' or ' Springfield.' 

Chowbent derives its name from a bent (A. -Sax. 
beofiet), or common, which was owned by one Chew or 

Croft. — A.-Sax., 'a small farm.' 

Cronton. — The name of this place occurs in the 
Testa de Nevill as Grohinton (fol. 396''), and as Crohmton 
(fol. 398^)^; but in 1562 we find a spelling — Crawenton — 
which (unless we have here the inevitable personal appella- 
tion) would imply that the original Saxon iihi, or farmstead, 
received its designation from a settlement of crows — A.-Sax. 
crawe (pi. crdwan), 'a crow.'^ 

Crosby. — The name of this Norse settlement was 
spelled Crosebi in Domesday, Crosshy in the Testa de Nevill ; 
and by 1645 it had reached its present stage (Crosby). This 
was 'the hamlet of the (stone) cross' — O. Nor. kross^ com- 
bined with O. Nor. bj'r, Scand. by, ' village,' ' town.' It is 
recorded in Baines ('Lane.,' 1870, ii. 396) that "An ancient 
cross stands in Little Crosby. This was formerly the object 
of a pleasing village festival called ' The Flowering of the 
Cross.' Mr. Nicholas Blundell, in an unpublished diary of 
the first quarter of last century, makes annual mention of it, 
as having attended it with his family." 

Crossens. — The earliest forms of the name, viz., 

^ Unpublished MS. Notes of Doming Rasbotham, written 1787 ; 
Baines, Lane, 1870, ii. 202. 

^ This difference in the initial letter is not here of much significance : 
the Romans at one time did not distinguish between the sound of C (k) 
and G hard ; but eventually the necessity of discrimination became 
evident, and G was formed by slightly modifying C. See the author's 
essay, ' A Fascinating Science,' in the Educational Times, January, 
1896, p. 35 ; and, for the vagaries of G in another tongue, his paper, 
'The Humours of a Great Language — Russian,' in the lournal oj 
Education, December, 1894, p. 710. 

^ Crdwe forms one of the select body of onomatopoetic or imitative 


Crossnes in the Scarisbrick deeds (thirteenth century), and 
Crosnes in the Nonarum Inquisitiones, fol. 40^ {temp. 
Edw. III.), point to but one etymology — O. Nor. kross, 
'cross,' and ?ies, 'a ness or shp of land,' in allusion to the 
comparatively elevated position occupied by the place 
under discussion between Martin Mere and the estuary of 
the Ribble. The Rev. W. T. Bulpit says that Archdeacon 
Clarke used to argue that Crossens meant Cross Sands ! 
The ancient cross of the township has been succeeded by a 

Croxteth. — In 1228 we find the spelling Croxstath. 
The latter element may be referred to the O. Nor. stathr 
(A. -Sax. stede, Ger, stadt), ' a stead,' ' a place '; while the 
first syllable is probably the O. Nor. krokr (Dan. krog, 
Swed. krok), which signified (i) anything crooked, (2) a nook 
or corner, (3) a personal name,^ as in the place-name 
Krdksfjorthr, which would make, in Mod. Eng., Croxforth. 

Cuerdley. — In the Survey of 1320-1346, Chetham Soc, 
vol. Ixxiv., 1 868, we find Cuerdesleglie mentioned as one 
of the members of the manor of ' Wydnesse.' The first 
element is no doubt a personal name, the second being 
from the common A.-Sax. leak (ley, lea), 'pasture-land.' 

Culcheth. — After careful study of the evidence which 
is so far available, I am unable to agree with the con- 
clusion come to by many, both locally and extraneously, 
that this place is the Cealchyth or Ccekhyth which is so fre- 
quently mentioned in Saxon charters, and was upon various 
occasions the scene of geniois, or meetings of the Supreme 
Council of the Anglo-Saxons, as well as of synods of their 
Church. It is true, as Mr. Robson, a local antiquary of 
considerable repute, pointed out,- that one of the farm- 
houses is, or was, moated round and called the ' Old Abbey '; 
but I do not see how we can get behind the cold fact that 
A.-Sax. Ceak-hythwiQ^.n's, ' the chalk hithe, wharf, or landing- 
place,' and that there is no chalk at Culcheth, nor any river. 
Besides, Bosworth^ and Thorpe,* as well as Toller,^ unite 

^ Krok the Peasant (Croc Agrestis) — Saxonis Grammatici Gesta 
Danorum, lib. viii., Holder's ed., i8S6. 

- Baines, Lane., 1870, ii. 21S. ^ A.-Sax. Diet., 1838. 

■* A.-Sax. Chronicle, 1861. 

' A.-Sax. Diet. (Bosworth and Toller), almost completed. 


in assigning (though Thorpe doubtfully) Cealchyth to 
Challock, or Chalk, in Kent. In early post-Conquest times 
Culcheth was styled Culchef, Kulchet, and Ctdchit. The 
natives of the place carry apocope still further, and, we are 
told, pronounce the name Kilsha. As to the etymology of 
the name, if we discard the idea of the common Celtic 
prefix Ml, i.e. cill (church) from Lat. cella, ' cell,' and of the 
Gael, coill, ' wood,' which usually gives the prefix cut or kil 
in Anglicized names,^ we may note that Cul is an Anglo- 
Saxon patronymic,^ as embodied in Culingworth or Culling- 
worth, while the second syllable of the name might be 
referable to the A. -Sax. cete (chete), 'cottage,' ' cell.' The 
form CulthetJi, occurring in the Nonarum Inquisitiones 
{temp. Edw. III.), fol. 40^^, suggests, however, a possible 
original A.-Sax. Ceald-hceth-, ' Cold Heath.' 

OunsCOUgh. — This name occurs in the Survey of 1320- 
1346, Chetham Soc, vol. Ixxiv., 1868, as Conescoughe. The 
meaning is ' rabbit grove ' — Mdle. Eng. coning, cuning, 
cunny, etc. (Scand. kani?i), through O. Fr. ultimately from 
hat. cunicu/us, 'cony,' 'rabbit'; and the dialectic survival 
of O. Nor. skogr (Swed. skog, Dan.-Norw. skov), ' grove,' 
' wood.' See Hargrave in the Wirral section. The cony 
or rabbit does not appear to have been introduced into 
England until after the advent of the Normans. 

Dallum. — ' In the dales ' — dative plural of A.-Sax. dcei, 
' dale.' 

Dalton, near Skelmersdale, is referred to in Domesday 
as Daltone — "the farmstead in the dale " — A.-Sax. dcel-tiin. 

Ditton. — The name was formerly spelled Dytlon. The 
English Dittons are generally traceable to A.-Sax. D'lciiin, ' a 
diked or ditched enclosure' — A.-Sax. die, masc. = 'dike,' 
fern. = ' ditch.' ' Dike ' and ' ditch ' are therefore doublets. 
In excavating a ditch a dike was necessarily thrown up. 
The name Ditton is consequently somewhat analogous to 
Walton where the latter refers to a wall or rampart and not 
to a weald, wold, or wood. 

Douglas. — The name of the river which flows through 
Wigan, and upon the banks of which, in contiguity to that 

^ Sir H. Maxwell, Scottish Land Names, 1894, p. 105. 
^ Birch's ed. of Kemble's Saxons in England, 1876, i. 450. 


town, tradition avers that King Arthur achieved several 
victories over the Saxons/ is derived from the Gael, dub/i 
(doo), 'black,' and i^/aise, g/as, 'water.' "This syllable 
gias,^' says Sir Herbert Maxwell,- " has two meanings ; as 
an adjective it means 'green ' or 'grey' [Wei. gias, 'blue,' 
' grey,' ' green '] probably cognate with the Lat. glauais ; 
as a substantive it means 'a stream.' Thus Dunglas is 
Gael, dim g/as, ' green hill' ; but Douglas is dud/i g/as, 
' the dark stream.' " Some have argued that ' Douglas ' 
means ' dark green,' ' water ' or ' river ' being understood ; 
but Joyce, in both his works on Irish topographic nomen- 
clature,^ gives ' Douglas ' as representing ' dark or black 
stream,' as above. 

DownhoUand, near Altcar, occurs in Domesday as 
Ho/and, wiiich is probably a corruption of the A.-Sax. /lo/t- 
/aiid, ' wood-land ' (Ger. Ho/z, ' wood ') ; / between two /'s 
would soon drop. There is a Holt Lane not very far from 
the township, which latter acquired the later prefix ' Down ' 
in order to distinguish it from Upholland (near Wigan), which 
lies high and contiguous to moors and woods. It was at 
one time thought that the name of the Dutch kingdom 
meant ' the low or hollow land ' (Dut. /lo/, A.-Sax. ho/, Ger. 
/w/i/, ' hollow ') ; but a ninth century document was dis- 
covered in which the spelling Ho /t/and occmxQd, and ' wood- 
land ' was as naturally applicable to the country as ' low land.' 

Eccleston. — A fairly common Lancashire ecclesiastical 
place-name, the equivalent of 'Churchtown' — ' Eccles ' 
coming through Late Latin, like Fr. eg/ise, Wei. eg/tvys, 

^ The fourteenth-century Monk of St. Werburgh's, Chester, Ranulf 
Higden, referring to Arthur's battles with the Saxons, says (Poly- 
chronicon, lil>. v.) there were four "super flumen Duglas . . . Ilodie 
fluvius ille vocatur AngHce Duggles, et currit sub urbe de Wygan, per 
decern miliaria a fluvio de Merseie distante in comitatu Lancastrice." 
Trevisa translated this (in 1387) "... uppon the ryver Douglas. . . . 
Now that ryver hatte [is called] Dugglys in Englische, and that ryver 
renneth under the citee of Wygan, that is ten myle from the ryver 
Mersea in Lancastreschire." — Polychronicon Ranulphi Higden Monachi 
Cestrensis, with Trevisa's and another translation, Babbington and 
Lumby's ed., vol. v. (1874), PP- 328, 329. 

- Scottish Land Names, 1894, p. 15. 

^ Irish Local Names Explained, 1S84, p. 40 ; Irish Names of Places, 
2nd sen, 1875, P- 266. 


etc., from Gr. eKKXrjcria, ' an assembly of citizens,' ' the 

Everton. — The earliest recorded mention of Everton 
seems to be in a document of 9 Hen. III. (1225),^ in which 
the name is spelled exactly as it is to-day. Everton was, 
however, probably one of the six unspecified berewicks 
noted in Domesday as pertaining to the manor of Derbei 
(West Derby). The late Sir J. A. Picton- followed Syers'^ in 
stating that Everton occurs in Domesday as Hiretun. But 
the context of the passage in Domesday shows that Everton 
cannot be intended, and that Beamont^ is probably right in 
assigning Hiretun to Tarleton. There would not at first 
sight appear to be a connection between the names of 
Everton and York, yet there is a fairly intimate one. The 
Saxons transformed theCelto-Roman^i^wrt«^;« or Evoracuin 
into Eofor-ivic or Efer-wic' (the variations in spelling are 
numerous, but the pronunciation remains practically the 
same), of which ' York ' is a corruption, the wic or wich 
(Lat. vicus, Gr. o?kos) being an Anglo-Saxon equivalent of 
' habitation,' ' village,' etc. Eofor or efer means ' a wild 
boar.' Everton was therefore originally the enclosure or 
farmstead of Efer or Eofor, which was a tolerably common 
Saxon name. It occurs, for instance, several times as a 
personal name in ' Beowulf,' " the oldest heroic poem in any 
Germanic tongue "; and, compounded, more than once in 
the famous list of benefactors to the Church of Durham, 
dating from the ninth century.*"' 

Farnworth. — 'The fern-farm.' A.-Sax, fearn (Ger. 
Earn), 'fern ' or 'brake '; A.-Sax. 7ueorthig, 'farm,' 'estate.' 

Fazakerley, we read, " was long the residence of an 
ancient family of that name," who, no doubt, derived their 
cognomen from the place. The origin of the first portion 
of the name seems obscure, but we can divide the word into 

^ Rotuli Litterarum Clausarum, ii. 64''. 

^ Memorials of Liverpool, 1875, '•• 334- 

3 Hist, of Everton, 1830, p. 41. 

* Domesday Cheshire and Lancashire, 1882, xv'\ 

® Anglo-Saxon /had the sound of w wherever possible. 

^ Liber Vitse Ecclesice Dunelmensis, Stevenson's ed., Surtees Soc. 
vol. xiii., 1841. Also included in Sweet's Oldest English Texts, 
Early Eng. Text Soc, 1885, pp. 153 sqg. 


three Anglo-Saxon vocables : fas, 'fringe'; aker (acer\ '(cul- 
tivated) field,' and kj (/ed/i), ' meadow,' which might imply 
' the meadowland with a certain or particular border or 
boundary field.' There is no other Fazakerlcy in England. 
The nearest approach to it is Fazeley, a place situate upon 
the border of the counties of Stafford and Warwick. We 
must reject the idea of a connection with the Ga.e\. /dsac/i, 
which, as an adjective, signifies 'desolate,' 'desert'; as a 
substantive, (i) 'grassy headland of a ploughed field,' 
(2) 'wilderness'; but it should be noted thvLt fatvs is a 
Northern dialect-word for ' fox,'i which has entered largely 
into personal and local nomenclature ; and, on the whole, it 
seems probable that this word forms the first element of 
' Fazakerley.' ' A(c)kerley ' by itself is an English local and 
personal name, and it probably had the same signification 
as 'Acre Dale,' which the 'Hist. Eng. Diet.,' quoting from 
Halliwell, defines as 'lands in a common field in which 
difterent proprietors held portions of greater or less quan- 
tities.'- I do not attach importance to an isolated case of 
'a(c)ker' being found in Cheshire to represent a sheet of 
water. ^ 

Fearnhead. — ' The high ground overgrown with ferns ' 
— A.-Sa.x. /earn, 'fern'; hedfod, 'head,' 'high ground.' 

Formby. — The name of this Scandinavian settlement is 
given as Fomebei in Domesday, and Forneby was the spelling 
in the thirteenth century. The mutation to ' Formby ' is 
simple, the later ;;/ resulting, phonetically, from the flat 

^ Halliwell, Diet, of Archaic and rrovincial Words, 1850. 

2 Much has of late years been written upon the medieval acre, and 
authors have differed. It may, however, be said roughly that the open 
or common field (A.-Sax. feld) was formerly divided into smaller fields, 
or shots (A.-S. scedt), or furlongs (A.-S. furh-lang, i.e., furrow-length) ; 
these, again, being split up into acres (A.-S. itcet-) of varying shapes 
and sizes, but perhaps most usually consisting of four or five selions 
(Lat. selio, Fr. silloji), or strips, or beds, separated from each other by 
balks (A.-S. bale), or ridges (A.-S. firycg), or riggs (O. Nor. hrygg\r)), 
or linches (A.-S. hlinc), or as they were, and are, often styled in the 
North, rains (O. Nor. rein, Ger. rain), i.e., unploughed strips. See 
Maitland, Domesday Book and Beyond, 1S97, pp. 373 sijtj.: Gomme, 
Village Community, 1890, pp. 193, 194; Seebohm, English Village 
Community, 1884, pp. 2, 106, 384, etc. 

3 Kendrick, Roman Remains Discovered at Wilderspool near 
Warrington, Transactions Chester Archceological See, vol. iii., 1SS5, 

P- 195- 


labial ^ immediately following the liquid n. In the same 
^■A.-^ Briinlmrgh became Bromborough, D2inb?'eatan, Dumbar- 
ton. ' Formby ' is thought locally to mean 'the pious or 
holy place.' It is true that the O. 'Nor. forn (long o) repre- 
sents ' an offering to God '; but I rather think that we must 
look for an explanation of the first element of the place- 
name to the O. Nor. /orn (short o), * old ' (A. -Sax. /yrn). 
Formby is the most prominent point on the South Lanca- 
shire coast, and would probably be the earliest settlement of 
the Northmen in that district ; and when other Norse posts 
were established this original one might easily be looked 
upon as ' the old village,' or Forn-by^ Saxon equivalents 
are Aldbrough, Oldham, etc. There is only one Formby in 
Great Britain. 

Unfortunately, Old Norse documents do not help us very 
much in regard to fixing the etymology of the less important 
Scandinavian settlements in England, especially those on 
the West Coast. Thus no Lancashire or Cheshire place- 
name is to be traced in the * Icelandic Sagas, etc., relating to 
the British Isles,' published by the Rolls Office ( text by 
Vigfusson, 1887; translation by Dasent, 1894), although we 
meet with Gri7ns-bcBr (Grimsby), Hjarta-pollr (Hartlepool), 
Hvita-byr (Whitby), Jbr-vik (York), Skardha-borg (Scar- 
borough), Jdrua-vibtha (Yarmouth), Man (Man), Onguh-ey 
(Anglesey), etc., in them; and this applies, with stronger 
reason, to Vigfusson and Powell's 'Corpus Poeticum Boreale,' 
1883, and the Danish History of Saxo Grammaticus ('Saxonis 
Grammatici Gesta Danorum,' Holder's ed., 1886), the first 
nine books of which History have been translated and edited 
by Elton and Powell respectively (1894). 

Freshfield is ' the field of the fresh-water spring or 
streamlet,' fresJi being an old dialect-word for a current of 
fresh water running into tidal water.^ 

Garston. — This name is referable to the A.-Sax. g(zrs^ 

1 Compare /£>r«^, 'first,' 'former,' 'fore,' in Halliwell's Diet, of 
Archaic and Provincial Words, 1850; and forne, 'anterior,' in the 
fifteenth-century vocabulary Promptorium Parvulorum, Way's ed., 
Camden .Soc, 1843-65 ; also the place-name /^?r«//a^z (Oldfield) in the 
celebrated Icelandic record, tlie Landnamabok. 

* " He shall drink nouglit but brine ; for I'll not show him 
Where the (\u\c)<^frc'shes are." — Shak., Tempest, iii. 2. 


* grass,' combined with the usual tihi. It denotes the use 
which the early Saxon proprietor made of the land. 

Gateacre.— 'The road-field '—Mdle. 'Eng. gate, 'way,' 

* road ' (compare Scand. gata (gade), ' way,' 'road,' ' street '; 
A.-Sax. geai), and acre {cecer), ' field.' I think, all things 
considered, that this is a more probable etymology than 
'goat-field' — A.-Sax. gdt, 'goat.' The name of Gates- 
head (-on-Tyne), which was long thought to mean ' Goat's 
Head,' is now concluded to represent ' Road's Head.' See 
Gavton in the Wirral Section. 

Glazebrook. — The village is named from the stream, 
which, like Alt and Douglas, bears a Celtic name — Gael. 
g/aise, glas, ' streamlet.' The invading Saxons added broc 
(brook) to the to them meaningless Erse term giaise. The 
form Glasebroc occurs in the Nonarum Inquisitiones, tcvip. 
Edw. HI., fol. 40. 

Golborne. — The early spelHngs of the name of this 
place, namely, Goldborne (1301), and Goldburii in the 
Testa de Nevill, perhaps indicated the yellow colour of the 
burn or brook whence the township derived its appellation. 

Haigh. — 'The hedged enclosure' — A.-Sax. haga. 

Hale.— In the thirteenth century we read of the " lands 
of Hales" and the "town of Halis." The name Hale is 
fairly common in England, and sometimes refers to a large 
residence or hall — A.-Sax. heall. We thus have an ancient 
Hale Hall here, and there is a Hale Hall in Cumberland ; 
but both ' hale ' and ' hall ' are frequently referable to 
A.-Sax. hcaKji) — O. Nor. hallir) — 'a slope,' pi. healas or 
halas ; and the early spellings quoted above would seem to 
lead to the conclusion that the original name of the Lanca- 
shire township was intended to signify ' the slopes.'^ 

Halsall. — This place is referred to in Domesday as 
Herleshala and Hekshale, while the present spelling Halsall 

^ In the Exchequer Lay Subsidy Roll for Lancashire, 1332, edited 
by Mr. J. Paul Rylands for the Miscellanies (vol. ii.) of the Lane, and 
Chesh. Rec'd Soc, it is surprising to find that, while only 16 names 
are returned for Manchester, with a total value of 46 shillings, the now 
insignificant village of Ilale stands with 25 names, and pays the sum of 
54 shillings. This Subsidy Roll practically forms a directory of all 
above the mere peasant class living in Lancashire in 1332. 


occurs in 1256. The first element in the name was once 
foisted by Canon Taylor,^ following a German fashion, 
upon the Celtic hal, ' salt.' The exact meaning of A.-Sax. 
health), which is generally rendered haia and hale in Domes- 
day, has been doubtful to the moderns. Until compara- 
tively recently everybody followed Kemble^ in his attribution 
of ' stone house,' ' hall '; but it is now accepted that ' slope ' 
or ' rising ground ' is the signification which is often meant, 
and this assumption is apparently borne out by the Old 
Norse word hall{r), ' slope,' ' hill.' In many English place- 
names with the termination under discussion the first 
element is a personal appellation, as that of the present 
Halsall doubtless is. There are no grounds for associating 
the A.-Sax. sele, O. Nor. sal{r) (dat. and accus. sal), ' hall,' 
' dwelling,' with Halsall. 

Haydock. — It is remarked by Baines or his editors^ 
that "this place is supposed to derive its name from the 
hedges of oak, or rather the oaks in the hedges, some of 
which, it is said, were planted as early as the reign of 
Edward the Confessor "; but this etymology is obviously 
impossible : it leaves unaccounted for the d which is present 
in all the older renderings of the unique name — Eydock, 
Haidoc, Haydok, etc. Taken literally the name would 
mean 'hedge-dock' — A. -Sax. hege-docce ; the hedges here- 
abouts may have been overrun with the troublesome dock 
or sorrel ; or pasture-land enclosed by a hedge — such 
enclosure being called in Anglo-Saxon a haga — may have 
been in this condition ; or Eydock may be simply the 
A.-Sax. eci-docce, ' water-dock.' In the absence of sufficiently 
early documental evidence — the place is not mentioned in 
Domesday or the Testa de Nevill — it is difficult to say 
with certainty what the modern ' Haydock ' originally repre- 
sented : there do not seem to be any grounds for connect- 
ing the name with A.-Sax. heg, ' hay.' But, on the whole, 
the latter element of the name is probably the E. Eng. doke 
or doak, ' a hollow '"* — O. Nor. dokk, ' pit,' ' pool ;' the whole 
name therefore representing ' the hollow place enclosed by a 

^ Words and Places, 1864, p. 392. 

'^ Codex Diplomaticus, iii. xxix. 

^ Lane, 1870, ii. 212. 

* Halliwell, Diet, of Archaic and Provincial Words, 1850. 


Hindley. — In the Testa de Nevill (fol. 406) we find 
Hindele. In the Survey of 13 20- 1346, Chethani Soc, 
vol. Ixxiv., 1868, the name occurs as Hyndekghe. It may have 
signified ' the hind or hinder meadowland '—A. -Sax. hindan 
and ledh ; but in view of other similarly-formed local names 
we must give the preference to an association with the 
female of the stag.^ As to ' hind,' a peasant, its Anglo- 
Saxon etymon hina, 'a servant,' invariably appears in place- 
names as hin-^ as in the numerous Hintons. There is still 
another possible, though in this case doubtful, connection : 
Kemble- makes a hind or hynd represent the third part of 
the hide, the Anglo-Saxon land-measure of varying acreage.-^ 

Hollinfare. — Literally ' the holly ferry ' {over the 
Mersey) — E. Eng. Jiolin (A.-Sax. /iok\g)n), ^hoWy' ; f am, 
'passage.' The contiguity of HoUin's Green, however, 
renders it possible that the ferry was attended to by a 
person named Hollin. 

Houghton. — ' The farmstead on the heel or projecting 
ridge of land ' — A.-Sax. hoh, 'heel,' 'ridge'; iu/i, 'farm- 

Hulme. — ' Low riparian meadowland ' — dialectic varia- 
tion of A.-Sax. holm. 

Huyton figured in Domesday as Hitune. The name 
means ' the elevated enclosure or farmstead ' — A. -Sax. hedh 
(Mdle. Eng. hig, hy, etc.), ' high,' combined with the usual 

Ince or Ince Blundell. — It seems almost natural to fly 
to the Celtic for the etymology of Ince (Wei. ynys, Gael. 
i?mis, Ir. ifiis, en?its, Scot, inch, ' island ') ; but where Domes- 
day, as in this case, has Hinne, and circumstances seem to 
preclude the theory of an island, or even of low riparian 
land, it is but reasonable that resort should be had to 

^ " The equine species has given to us ' Horsley '; the bovine, 
'Cowley,' ' Kinley,' and 'Oxley'; the deer, 'Hartley,' 'Rowley,' 
' Buckley ' and Hindley ; the fox, ' Foxley '; the hare, ' Harley '; and the 
sheep, 'Shipley.'" — Bardsley, English Surnames, 5th ed., 1S97, p. 119. 

- Codex Diplomaticus, iii. xxx. 

3 As Domesday units, Mr. Round (Feudal England, 1895, P- 37) 
gives 30 acres in the virgate, and 4 virgates in the hide, which there- 
fore represents 120 acres, 



another language. In Old Norse we have mni, ' inn,' 
'abode,' 'house,' 'hall' (there exists an Ince Hall). The 
genit. sing, of t'/mz is mm's, which would be used with 
dative significance with the preposition ///, ' to,' which the 
genius of the Icelandic language requires should take the 
genitive case, as in the phrase ok til sama in?iis, ' and to 
the same house.' Ines was the spelling in the reign of 
Edward III. {1327-1377). Blundell is the patronymic of 
the owners of the manor. 

Kenyon. — In the Testa de Nevill (fol. 405^) we find 
mention of Robert le Kenien, from whom this township 
derives its name. In the Survey of 1 320-1 346, Chetham 
Soc, vol. Ixxiv., 1868, we have the ?,]it\\mg Ke?iea?i. It is not 
quite clear what ' Kenien ' or ' Kenean ' means. It may be 
a mutated form of the O. Fr. chanoifie^ ' a canon ' ; or pos- 
sibly of O. Fr. kienin {chienin), ' doglike,' ' canine,' or O. Fr. 
kenon, ' little dog '; compare such names as William le Chien, 
Eborard le Ken, Thomas le Chene (O. Fr. c/ien (ken), ' dog '), 
quoted from mediaeval rolls, as nicknames from the animal 
kingdom, by Bardsley in his 'English Surnames,' 5th ed., 
1897, pp. 492, 534, 567. 

Kirkby. — The Domesday Cherchebi is the Norse equiva- 
lent of the modern English Churchtown — O. Nor. kirkja 
(Dan.-Norw. kirke^ Swed. kyrkd)^ ' church ' ; O. Nor. byr 
(Scand. by), ' settlement,' * village.' 

Kirkdale, the Domesday Chirc/icdek, is the low ground 
upon which stood the Northmen's church — O. Nor. kirkja 
(Dan.-Norw. kirke, Swed. kyrka), 'church'; O. Nor. da/r 
(Scand. da/), 'dale.' 

Knowsley is a corruption of ' Kenulf's Ley ' — A.-Sax. 
/eii/i, ' mcadowland ' — as the Domesday spelling Chenulveslei 
shows. Kenulf— A.-Sax. Chie IVulf, 'Bold Wolf.' 

Lathom. — This township and ancient chapelry was 
noted for many centuries as the seat of the Stanleys. The 
Domesday spelling of the name was Latune. In 122 1 
we find Ladhun ; and not very long afterwards {temp. 
Edward I.) we read of Sir Robert Lathom and "Thomas de 
Lathum his grandson." We have here the O. Nor. hlath 
or hlatha (dat. pi. hlathum) — Dan.-Norw. lade, Swed. lada — 


' Storehouse,' ' barn.' Lai/A or /af/ie is a Lancashire dialect- 
word for 'barn.'i 

Leigh. — We have a historical record that the leighs or 
leas (A.-Sax. /ea/i, ' meadowland ') in this district are, or 
were, '* green and luxuriant."- It is said that the original 
guttural pronunciation of Leigh is still retained by the 
natives. West Leigh and Astley (East Leigh) are situated 
with respect to Leigh exactly as their names imply. 

Linacre represents ' the lint or flax field ' — A.-Sax. /hi, 
' flax,' ' linen,' and acer, ' field ' ; O. Nor. lin-akr, ' flax field.' 

Litherland occurs in Domesday both as Liderlant and 
Literland. In the time of King John (1199-1216) we find 
Litterland ; of Edward II. (1307-1327), when Norman- 
French linguistic influence was being considerably weakened, 
Lytherland ; and in 137 1, Letherland. The probable etymo- 
logy is the O. Nor. hruliar-la7id, ' the sloping land,' which 
corresponds exactly with the natural aspect of the township, 
liut see under Liverpool. 

Lowton was formerly Lauto7i,^ i.e., the tun or settle- 
ment by or on a mound or rising ground — A.-Sax. hlcew, 
'mound,' 'tumulus,' 'hill.' This word {hlcew) has come 
down to us in place-names both as loiv and laia, the latter 
spelling and pronunciation now being mainly confined to 
the South of Scotland.^ 

^ I'icton, South Lancashire Dialect, Proceedings Lit. and Phil. Soc. 
L'pooJ, vol. xix. (1864-65), p. 35. 

- Baines, Lane, 1870, ii. 203. 

^ Survey of 1320- 1346, Chatham Soc, vol. Ixxiv,, 1868. 

* This vowel-change is not restricted to Old English, "^m is the 
only diphthong which the Latin language has preserved, that is, in the 
generality of cases ; for here also we find a weakening — to — common 
in early times. It is observable, however, that the new form in never 
drove out the old one in au ; but the two remained side by side. . . . 
Corssen supposes that au was employed by educated men in words 
where was heard in the mouth of the countryman. This is borne out 
by the anecdote of Suetonius ( Vespasian, 22) which Corssen quotes. 
The homely Emperor was taken to task by the courtier Florus for 
calling a plauslruDi (cart) a plostruvi ; and retaliated next day by 
pronouncing his critic's name as befitted ears so polite — Flaurus. . . . 
Somewhat analogous to the change of sound from au to in Latin is 
the pronunciation of au in French ; and in some parts of tlie North of 
England 'law' is pronounced like lo." — Peile, Greek and Latin 
Etymology, 1869, pp. 154-155. 



Lunt or Z?oid here, as in other Danish districts, is 
referable to the O. Nor. htndir), ' a sacred grove.' Lunt 
was therefore the scene of the pagan rites of the ancient 
Scandinavians who settled in the region. 

Lydiate. — This name occurs in Domesday as Leiaie, 
which may have meant originally ' the road over the 
meadow' — A.-Sax. Icd/i, 'lea,' 'meadow'; and geat, 'gate,' 
'road,' 'passage'; but it is more probable that the name is 
simply the A.-Sax. hlidgeat (Mdle. Eng. lidgate or lidyate), 
' lidgate,' 'postern.' The word is found in some rural 
districts as ' lidgett ' (which is simply a slurred pronuncia- 
tion of ' lidyate ' or ' lydiate '), and in London it appears as 
Ludgate ; in the former case the old palatal pronunciation 
of g (as in ' young ' and ' year,' A.-Sax. geoiig and gear) 
being retained, in the latter the guttural (hard) enunciation 
surviving. We find the spelling Lydyate in the Nonarum 
Inquisitiones [temp. Edvv. IIL), fol. 40''. 

MaghulL — Domesday gives Magele as being one of six 
manors held by the Saxon Uctred before the Conquest. In 
the Testa de Nevill (fol. 396) the name occurs as Maghale. 
The present spelling is found in 1635. Some have argued 
(urging the traces of strong Irish influence in the neighbour- 
hood) that the first element of ' Maghull ' must be the Gael. 
iiiag/i, ' a plain '; but there is little doubt that the magh in 
' Maghull ' is no more of Erse origin than is the equally 
Gaelic-looking ai/gh in 'Aughton,' and that the name 
should really be divided — taking the Testa de Nevill 
spelling as affording the best clue to the etymology — as 
follows : Alag-hale, where hale (A.-Sax. healh, O. Nor. hallr) 
is the common topographic suffix meaning ' a slope.' Mag- 
hull is situate on the only rising ground in the immediate 
neighbourhood. A personal appellation is frequently found 
prefixed to English place-names ending with mutated forms 
of liealh or hallr, but in such cases we invariably have 
clear traces of the genitive (possessive) s or n, which is 
entirely lacking in ' Maghull.' A connection with the 
common A.-Sax. vmg or niaga, 'son,' 'descendant,' 'kins- 
man,' is, however, within the bounds of possibility. 

Although, as I have said, I discard the notion of the name 
Maghull being of Erse origin, it is nevertheless interesting 


to note that the place was " formerly often called Mai7 or 
Ma/e,"^ a circumstance which is in agreement with Irish 
phonetics, in which g/i, under certain conditions, is mute. 
The spelling ATael, too, occurs in a well-known Lancashire 
record of the first half of the fourteenth century.- Now 
when we recollect that inael is Old Irish^ for ' a bald or bare 
hill,'' it seems hard to resist the inference that ' MaghuU ' is 
of Erse origin. But it is quite clear that, while a Alacl 
might result from a MaghuU^ a MaghuU could not possibly 
result from an original Mnel. The old Irish plantation 
influence in the district, already mentioned, was evidently 
responsible for the Hibernicising of the Magele of Domes- 
day and the Maghale of the Testa de Nevill into Mael. 

Makerfield. — The first portion of this name has much 
tested the resources of etymologists, and little wonder, con- 
sidering the variation in spelling which it has undergone in 
the course of its career — Macerfcld, Maserfelth, Maresfeid, 
Maxsefeld, etc. An unlikely derivation put forward a long 
time ago was from the Gael, magh, 'a plain.' I do not 
remember to have seen the A.-Sax. mceger (O. Nor. magr^ 
Lat. macer, Eng. meagre), 'lean,' 'poor,' 'barren,' advanced. 
The Swedes have the expressions mager Jord, 'poor or barren 
soil,' and viager aker, ' a sterile field '; while Cicero con- 
nected macer with sohan, indicating ' poor soil.' We must 
recollect that the A,-Sax. and Ger. /e'/^/ originally denoted a 
place where trees had been felled. Local antiquaries 
identify Makerfield with the Maserfelth where Oswald, king 
of the Northumbrians, was slain in battle with Penda, king 
of the Mercians, in 642, as related by the Venerable Bede ; 
but this is not allowed in general history. Stevenson, refer- 
ring to the conflict, says : " Although a place near Winwick, 
in Lancashire, named Mascrfield, has claims to be regarded 
as the spot where this battle was fought, yet there are much 
stronger arguments in favour of Oswestry {i.e., 'Oswald's 
Tree '), in Shropshire.'"^ And Plummcr, the latest editor of 

^ Baines, Lane, 1870, ii. 425. 

- Survey of 1320-1346, Chclham Roc, vol. Ixxiv., 1868. 
^ Modern Irish and Gaelic have the form uiaol (Welsh, tiioel), ao 
being a modern aphthong substituted for the older ae and oe, 
■• Joyce, Irish Local Names Explained, 1884, p. 104. 
^ Bede's Historia Ecclesiastica, Stevenson's ed., 1S41, i. 177. 


Bede, equating Maserfelth with Oswestry, makes no mention 
of the claims of the Lancashire Makerfield.^ The student 
of battle-sites should not, however, omit to note Mr. Browne's 
'Pre-Norman Sculptured Stones in Lancashire,' in the 
Transactiojis of the Lancashire and Cheshire Antiquarian 
Society, vol. v., 1887. The Welsh name of Oswestry is 
Croes [cross] Oswallt. 

Melling. — The Domesday spelling is Melinge. At first 
sight we would appear to have here a pure Anglo-Saxon 
patronymic ; but there is little doubt that ' Melling ' is a 
Saxonisation of the name of the Norman, Vivian de Molines, 
to whom, shortly after the Conquest, Roger de Poictou 
granted a large tract of land in this district. The Molyneux 
family trace their descent from William des Molines, so 
named from Moulins (' the Mills '), a town of Bourbonnais 
in France. The historical gradation in the spelling of this 
family name is as follows : Molines, Molynes, Moulins, 
Mulans, Mulynes, Mulyneus, Molineux, Molyneux. 

Mersey. — The circumstance that the Mersey, unlike 
the Dee, the Severn, the Thames, and other English rivers, 
does not seem to bear a Celtic name, was once thought to 
confirm a widely-accepted theory that within historic times 
the Dee and the Mersey had but one estuary between them, 
and that certain defined depressions of the Wirral peninsula 
once conveyed the superfluous waters of the Mersey into 
the then larger volume of the sister river. The Belisama 
of Ptolemy, after a protracted controversy, was concluded 
to be the Ribble, and, there apparently being a lack of 
reference to the Mersey during the Roman sway in this 
country, it was estimated, seeing that the Danes and Norse- 
men sailed up the river, that the Mersey must have cut 
itself an independent outlet to the sea, by way of what is 
now Liverpool, between the fifth and eighth centuries. But 
Mr, T. G. Rylands, F.S.A., who, originally instigated by a 
friend's interrogation as to whether the Belisama was the 
Mersey or the Ribble (the Editors of the ' Monumenta 
Historica Britannica,' 1848, gave both), devoted almost a 

' Bede's Historia Ecclesiastica, Plummer's ed., 1896, i. 451, ii. 494. 


lifetime to the study of Ptolemy, has conclusively proved^ 
that the Belisama can only be the Mersey. 

Mr. Rylands's researches into Ptolemy's geography of our 
north-western coast may be summed up as follows : 

langanorum Prom. = BrachypwlL 

Tisobius = Traeth Mawr. 

Seteia = Dee. 

Belisama = Mersey, 

Setantiorum Portus = Ribble. 

Ituna = Solway. 

In Domesday the Mersey figures as Mersha (' Inter 
Ripam et Mersham '). The earliest Saxon document 
mentioning the Mersey which has so far been found is the 
will of Wulfric Spott, which was confirmed by King 
Ethelred in 1004. This Wulfric was a Mercian nobleman, 
a large landowner (even for those times), and the founder 
of the Abbey of Burton-on-Trent. He bequeathed to his 
sons ^Ifhelm and Wulfag, ititer alia, the lands "betwux 
Ribbel and Mcerse and on Wirhalum." The will was 
printed by Sir William Dugdale,- with a Latin translation 
and an identification-list of the place-names mentioned in 
it. Sir Francis Palgrave^ describes the testament as "a 
singular and important document requiring much topo- 
graphical and legal illustration." He adds that " Dugdale's 
translation is not particularly accurate," a term which might 
also be applied to Sir William's transcription of the will. 

It is worth remembering that just as we have a Wallasea 
(island) on the coast of Essex, so have we a Mersea^ (island) 

^ Ptolemy Elucidated, 1893, passim. The broad results of his 
elaborate investigations had, however, long previously been published, 
viz., in the Transactions Hist. Soc. Lane, and Chesh. : Ptolemy's 
Geography of the Coast from Carnarvon to Cumberland, 1877-78, 
pp. 81 sqq., and The Map-History of the Coast from the Dee to the 
Duddon, 1878-79, pp. 83 sqq. 

^ Monasticon Anglicanum, 1682, i. 266 ; ed. 1S17-30, iii. 37. 

* Rise and Progress of the English Commonwealth : Anglo-Saxon 
Period, 1832, ii. 293. 

■* This is the island mentioned in the entry in the Saxon Chronicle 
under A. D. 895, which describes how, in this year, the Danish army 
departed from Wirheal (Wirral) into N. Wales, marching thence 
across Northumbria and East Anglia (striving to avoid Alfred's forces) 
into Essex, finally taking up their quarters on Mersea — " oth thast hie 


there. Mcrsca (A.-Sax. Meres-tg) has been explained as 
' the sea island,' and Mersey, by analogy, as ' the sea river.' 
The latter interpretation, at least, is illogical. The A.-Sax. 
f//ere (i) 'sea' (2) 'lake,' 'pool,' 'niere/^ there is little 
doubt, is the base of both names ; but, in my opinion, it 
must here be translated ' pool ' or ' lake,' not ' sea,' as, for 
example, in Alfred's interpolation of 'The Voyages of 
Ohthere and Wulfstan ' in his translation of Orosius, where 
he speaks of " swi'the micle meras " (very large meres) 
throughout " tha moras " (the moors). The low-lying lands 
at the mouth of and in the estuary of the Mersey, have, 
from time immemorial, like the flat coast of Essex, been 
subject to inundations, with the natural result of the forma- 
tion of meres and marshes (A.-Sax. f/iersc, for merisc, lit. 
' mere-ish '). And I believe the s in both ' Mersey ' and 
' Mersea ' may represent not the genitive singular but the 
nominative plural case, which, although (as we have just 
seen) as in the classic West Saxon, was, like the gen. sing., 
es in the Dano-Saxon dialects. ' Mersey ' would therefore 
correspond to an original Meres-ed, just as ' Mersea ' repre- 
sents an original Meres-'ig — A.-Sax. ed, O. Nor a, ' river,' 
'water;' A.-Sax. f^(form o^ed), O. Nor, ey, ' island.' It will be 
seen that to be quite regular the Essex Mersea and the Lanca- 
shire and Cheshire Mersey should be reversed in spelling. 

Whitaker, the historian of Manchester, 120 years ago, 
was within reasonable distance of what we must accept as 
the true etymology of ' Mersey.' Reading Mcersc (instead 
of Mtcrse) as the name of the river in the will of Wulfric, 
he remarks : " from the marshes and marshy meadows that 
skirt its channel on both sides, in one continued line to the 

comon on Eastseaxna land easteweard on an it^land thret is iite on 
thwre s:e, thxt is Meresighiitxi." This part of the Chronicle dealing 
with Alfred's campaigns against the Danes Dr. Sweet describes (Anglo- 
Saxon Reader, 1894, p. 35) as " a perfect model of Old English prose," 
while Prof. Earle considers it to be "the most remarkable piece of 
writing in the whole series of Chronicles." In Elhelwerd's Chronicle 
{lib. iiii.) the Essex Mersea is referred to as "a place in Kent "! — "ad 
Meresige locum Cantiam " {sic). 

^ The A.-Sax. jno-e has not come down to our day in its original 
form without undergoing metaplasm in the inleival. For example, 
Levins's Manipulus Vocabulorum, 1570 (printed by the Early Eng. 
Text Soc. ), has the epenthetic form meare, 

• 1' 


sea, obtaining the descriptive denomination of Mersc-ey, 
Mers-cy, or marshy water. "^ 

A word may be added as to the conceivabiHty of the first 
element of ' Mersey ' being a rehc of Celtic nomenclature. 
At first sight there does not appear to be much in common 
between Belisama and Mersey ; but a phonological exami- 
nation of the name which Ptolemy probably wrote down 
from the lips of Roman or Greek sailors may perhaps induce 
us to take a contrary view. The most noticeable point to 
begin with is the occurrence of s in the middle of both 
Belisama and Mersey. Turning then to the first letter of 
Ptolemy's name, l>, phonologists well know that this is 
readily mutable to both v and ;//, and conversely.- £ is the 
second letter in both names. Then we have / in Belisama 
and r in Mersey. This presents no difficulty ; / and r, both 
liquids, are or have been interchangeable in many languages, 
the greater tendency being for r to slide into the easier- 
sounded /.^ The dual occurrence of medial s having 
already been noted, this, of course, is as far as the com- 
parison can be taken. The apparent similarities may only 
be coincidental, but it is at any rate just conceivable that 
the Saxons might transform, say, a Celtic Marusia (' dead 
water '), a name which possessed no meaning for them, 
into Mcres-eci, which was full of significance to them. But, 
after all, this is pure conjecture. 

From a work entirely devoted to the Mersey, like Dr. 
Blower's,* we might have expected some treatment of the 

^ Hist, of Manchester, 1775, ii. 23S. 

- See Peile, Greek and Latin Etymology, 1869, pp. 234, 235, and 
Ma.K Miiller, Science of Language, 2nd sen, 1864, p. 145 ; com- 
pare the Welsh mutations — e.g. luira, J bread'; dy fara (vara), 'thy 
bread '; fy niara, ' my bread '; the Irish d, /J/ = vov -w ; Mod. Gr. ^ — v, 
Lat. Sahrma = Severn, Lat. Abona = Avon, etc. ; and note, in our 
own language, 'somersault,' from Fr. soubresaut ; 'malmsey,' from 
Fr. malvoisie ; 'marble,' prim, from Lat. juar/nor. 

3 Peile, Greek and Latin Etymology, 1S69, p. 81 ; Max Miiller, 
Science of Language, 2nd ser., 1864, jip. 165, 166; Morris, Historical 
English Grammar, 1875, P- 44) etc. Eng. 'palfrey ' is ultimately from 
Low Lat. paravercJiis ; 'chapter,' conversely, from Lat. capilitluin ; 
' Gibraltar ' is a corruption of Gibel-al- Tarik. The Chinaman is perhaps 
the greatest adult sufferer from lalhtion or In his mouth 
' America ' becomes Melika. 

■* Mersey, Ancient and Modern, by Benj. Blower, 1878. 



name, but I can find nothing in the book beyond a Hteral 
reproduction from Baines of an untenable etymological 
connection with ' Mercia,' which Sir Peter Leycester seems 
to have suggested originally : speaking of the great Midland 
kingdom this Cestrian chronicler said •} " It was called 
Mercia, not from the river Mersey running from the corner 
of AVirral, in Cheshire, because that river was the utmost 
limit thereof westward ; but I rather believe that river took 
denomination from this kingdom, which it bounded on that 
side, and was called Mercia, because it abutted or bordered 
upon part of all or most of the other kingdoms of the 

Netherton. — Anglo-Saxon : ' the lower enclosure or 
farmstead ' — A.-Sax. neothor^ ' lower.' 

Newsham. — This name now survives only in Newsham 
Park and Newsham House. In the Testa de Nevill it is 
spelt Neiisti7)i (fol. 403) and Neusom (fol. 409), forms which 
point to an original A.-Sax. Niwehustwi, ' at the new 
houses,' husiim being the dative plural of Mis, 'house.' 
Other English Newshams figure in Domesday as Newe- 
husufi, Newhuson, etc. \ and the Durham Newsham appears 
in the Boldon Book as Newsom. 

Newton is the A^eweton of Domesday — A.-Sax, 7ihve, 
' new.' The primary signification of A.-Sax. f/m was ' enclo- 
sure'; then it naturally came to mean 'farm' or 'manor'; and 
the nsLxnef/m was retained when, in process of time, a village 
grew up round the farm. There are several scores of 
Newtons in England, and it is probable that our Newton, 
like most of the rest, simply denotes the spot where a Saxon 
cultivator had taken up a new estate ; but Dr. Robson, the 
Warrington antiquary, had a much less prosaic theory of the 
origin of the name. He says that, on the death of King 
Oswald, the traditionary site of whose palace, and of the well, 
is at Wood End, near Hermitage Green, " the royal residence 
seems to have been transferred to another site to which 
naturally enough the name of Newton, the new town or 
vi7/, was given."- 

' Antiquities of Cheshire, 1673, p. 92. 

2 Transactions Hist. Soc. Lane, and Chesh., 1851-52, p. 205. 


North Meols. — In Domesday we have A/ek. This 
name is referable to the O. Nor. me/r (genit. ??ie/s, pi. me/ar), 
'sand-hill,' ' sand-bank.' 

Orford.— * The cattle-ford '— A.-Sax. or/, ' cattle ' ; and 

Ormskirk does not figure in Domesday, but the first 
settled possession of the ancient parish, and the establish- 
ment of the kirk, are ascribed by local historians to a 
Northman bearing the common name Orm — O. Nor. or//ir, 

Orrell. — ' Orrell ' is all that has come down to us of the 
name of a place which in Domesday is chronicled as 
Oies^riinck. In the last element of the ancient designation 
we have, of course, the O. Nor. mclr (pi. vielar, genit. sing. 
mels), * a stretch of sand.' O and U are the Scandinavian 
negative particles. Thus, in Swedish, tack is ' agreeable,' 
otlick is ' nasty.' It is probable that otdck or its etymon has 
nothing to do with Otegri ; but there is an O. Nor. word 
teigr, 'a strip of field or meadow-land'; and, as Cleasby 
(' Icelandic Diet.') has pointed out, the use of the Icelandic 
negative prefix // or b is almost unlimited. The hypothesis, 
therefore — I give it for lack of a better — is that Orrell repre- 
sented roughly, originally, ' the sandy area out of which not 
a strip of meadow-land could be got.' Another Orrell, in 
Sefton parish, is set down in Domesday as Otrbigemele. 

Padgate. — 'The path-gate ' — A.-Sax, pce^-geat. 

Parr. — This name was formerly spelled Par re. A parr 
(etymology dubious) is a young salmon ; but it is of course 
impossible to connect this fish with Parr. The name is 
probably a diminution of A.-Sax. parrtic or pearroc (Scot. 
parrok, Eng. parrock), 'croft,' 'enclosure,' 'park.' Compare 
Mdle. Eng. parren, ' to enclose ' ; parred, 'enclosed.' This 
etymology would seem to be confirmed by a note in Mr. 
Way's edition of a celebrated 15th cent, vocabulary^ under 
parrok — viz : " In Norfolk, according to Forby, an enclosed 
place for domestic anii.ials, as calves, is called a par, and 
the farmyard, containing pars for the various animals which 
inhabit it, is called a par-yard." 

1 Promptorium Parvulorum, Camden Soc, ii. (1S53) 3S4. 


Pemberton. — There is no name resembling Pemberton 
to be found in the Testa de Nevill ; but the place is men- 
tioned, spelt as it is to-day, in the Survey of 1320-1346, 
Chatham Soc., vol. Ixxiv., 1868. Despite the scantiness of 
early information as to the name — a fault not uncommon 
with Lancashire place-appellations compared with those of 
the more fertile, earlier-settled, and consequendy anciently 
more thickly peopled counties of the South, East, and West 
of England — there is little difficulty in coming to the con- 
clusion that the original Anglo-Saxon name was Pin-bearu- 
tihi, ' the pine-grove farmstead.' Compare Pamber Forest 
in the South of England. 

Penketh. — This name has apparently suffered much 
curtailment and assimilation in the course of centuries, and 
little beyond a more or less reasonable guess can be made 
as to its full original meaning. The earliest spelling recorded 
seems to be that in the Testa de Nevill (fol. 396), viz., Fe?i- 
ket, where the dropping of the present-day final // does not 
necessarily mean that the name was pronounced as spelt by 
any but Norman-French scribes. 

Penketh, however, occurs in the Survey of 1320- 1346, 
Chetham Soc, vol. Ixxiv., 1868. The last three letters in 
Penketh may represent, as they do in Lambeth (Lamb-hithe) 
the Anglo-Saxon hyfh, ' hithe,' ' haven,' ' landing-place ' -^ 
while the first element of the name may be a mutilated 
patronymic, or even refer to a (silver) penny rental — A.-Sax. 
penifig (penny). We can imagine how easily Peii{ii)ing-hyth 
(or the genit. pi. Pen{n)inga-hyth) could eventually be cor- 
rupted into ' Penketh.' On the other hand, as there is a 
heath close to the township (at any rate, there is one marked 
on the last Ordnance map of the district) it is not impossible 
that eth may represent the A.-Sax. hceth (heath). There are 
no physiographical or general grounds for Canon Taylor's 
classification ('Words and Places,' 3rd ed., 1873, p. 147) 
of Penketh with place-names associated with the Welsh pen, 

A Cymric origin can, however, it may be interesting to 
note for purposes of comparison, be claimed for Penkridge 

1 This might seem to be borne out by the spelling Penkythe, which 
is found in the Calendarium Inquisitionum Post Mortem, ii. {temp. 
Edw. Ill,) 238. 


in Staffordshire. The ancient name was Fennocruciuvi, 
which Professor Rhys, perhaps our chief Celticist, explains 
as follows •} " penno-s, ' head ' or ' top ' ; cnlcio, which 
became in ^V'clsh crFic, now cnlg, a ' heap ' or ' mound ' ; 
the whole would mean the top or head of the mound or 

Pennington was the tun or settlement of the Saxon 
Penning family.- Occasionally, however, a place-name with 
Feiining may have reference to a rental. Thus Penning- 
hamc, near Newton-Stewart, is said to be a ham or piece of 
land which was held at a charge of a silver penny (A. -Sax. 

Poulton. — ' The farmstead by the pool ' — A.-Sax. pul, 
'pool'; tun, 'farmstead,' 'manor.' 

Prescot. — ' The priest's dwelling ' — A.-Sax. preost, 
' priest,' and cof, ' cottage.' According to tradition Prescot 
" was anciently the habitation of priests." 

Rainford is a corruption of Randleford, Randle being 
the name of a brook running through the village.^ 

Rainllill. — In the Survey of 1320-1346, Chetham Soc, 
vol. Ixxiv., 1868, we have the spelling Koynhidl. In the early 
part of the 15th century Raynliill occurs. The name may 
be a corruption of A.-Sax. hrcrfn-hyll, ' raven-hill,' or pos- 
sibly an assimilation of some personal appellation (Rain- 
ham in Kent was originally Roegingaham) ; but the 
former theory is the more feasible. The raven was the 
war-emblem of the ancient Danes, and Raven Hill, in 
Whitby parish, N. Yorks, obtained its name from the 
Danes having set up their standard upon it after landing 
in 867. There is, moreover, a village called Ravenhead 
within a short distance of Rainhill. But, on the whole, I 
am inclined to think that the first element of the name is 
nothing more than the Northern dialect-word rain, ' ridge,' 
'balk,' etc. — O. Nor. rein {cf. Gcr. rain) 'strip of land ' — in 

1 Celtic Britain, 1884, p. 303. 

2 Birch's ed. (1876) of Kemble's Saxons in England, i. 470. 
' Rev. J. Bridger to the author. 


allusion to the old method of cultivating the hill, viz., sepa- 
rating the ploughed portions by rains, i.e., ridges or balks.^ 

Ravensmeols. — The name of this Scandinavian settle- 
ment occurred in Domesday as Erengermeles, the latter por- 
tion of which is the usual O. Nor. inelr (genit. case mels), 
pi. melar, ' sand-hills ' or ' sand-banks ' ; while the front part 
of the name may be referable to the O. Nor. eyrr (Dan. ore, 
Swed. or) ' a gravel bank ' (of a river or of small tongues of 
land running into the sea), and the O. Nor. eng or engi, 
pi. engiar, 'meadows,' the name, upon this hypothesis, 
meaning ' the gravel meadow strips among the banks or hills 
of sand.' This is, of course, presuming that the name 
recorded in the Norman Survey is an approximately correct 
rendering of the native appellation, which may otherwise 
have embodied a personal name, the raven (O. Nor. hrafti, 
Dan. ravn) being the war-emblem of the ancient Danes, 

Risley was 'the brushwood pasture' — A.-Sax. hris, 
'brushwood'; kah, 'pasture'; or, perhaps, as Risley 
borders upon the large moss of that name, 'the rushy 
meadow ' — A.-Sax. rise, ' rush.' 

Rixton. — The spelling is the same in the Testa de 
Nevill. This was ' the farmstead by the rushes ' — A.-Sax. 
rix, rise, 'rush ' ; iihi, ' farmstead,' ' manor.' 

Roby. — Domesday spelling J^a/fii. The name of this 
village is practically identical with that of Raby on the 
opposite side of the Mersey, the Rabiloi the Norman Survey 
" Inter Ripam et Mersham " doubtless being a copyist's 
mistake for Rabic," the spelling which figures in the Cheshire 
Domesday. In both ' Roby ' and ' Raby ' we probably 
have, prefixed to the usual Scand. word for 'settlement,' 
' village ' — by — the animal-name of the original Scandinavian 
owner of the settlement, the O. Nor. rd, pron. as roiv^ 
noise — Dan.-Norw. raa, Swed. ra, both pron. raw — mean- 
ing 'a roe.' This animal-name was borne by more than 
one member of the royal family of ancient Denmark. 

J See Note, p. 45 ; also Halliwell, Diet, of Archaic and Provincial 
English, 1850 ; and Seebohm, English Village Community, 1884, 
p. 381. 

2 Raby occurs in The Great de Lacy Inquisition of February 16, 
131 1, Chetham Soc, vol. Ixxiv., 186S. 


Sankey. — Some of the old spellings found of this name 
are : Sa/iki, Sanky, Sanchi\ Sonky, Sankye, Sonkey, Sanckey, 
and Sonkie. All these seem to point to one signification : 
* the sunken (or low) place by the water ' — A. -Sax. sencan 
(pret. sing, sanc)^ 'to sink'; ig (/y) primarily 'island,' but 
also indicating low riparian land. 

Mr. Beamont^ mentions the absurd derivation " from the 
words ' sand ' and ' quay ' " ; and I can trace no historical 
grounds for his alternative etymology, " the Sank, the brook 
which goes through the place, with the ey or eyot at its 

In some ' Notes on the Local, Natural, and Geological 
History of Rainhill,'^ by the Rev. H. H. Higgins, M.A., 
occurs the following passage : " Rainhill was then part of an 
island, or rather of a peninsula ; all the flat lands of Speke 
and Ditton were under the sea, which swept round far to the 
east of Rainhill, leaving Appleton and Bold high and dry, 
but pouring its waters over the Sankey Brook district. . . . 
This was the condition of things probably not so very long 
ago, I mean since the appearance of man upon the earth, 
though how many thousands of years ago it would be a mere 
guess to suggest." 

Scarisbrick. — In an Inspeximus of 17 Edward II. 
(1324) contained in the chartulary of Burscough, a charter 
without date is cited in which occurs the name "Walter, 
lord of Scaresbrek." We meet with Scaresbreck in 1508, and 
with Scarisbrick about the middle of the 1 7th century. The 
ancient spellings imply ' the slope-land belonging to one 
Scar or Skardh ' — O. Nor. brekka, ' a slope.' As a personal 
appellation the O. Nor. skardh, ' a gap or cut in a rock,' 
meant ' hare-lip.' Skardh or Skardhi was a frequent 
Danish proper name on the Runic stones (' Scarborough ' is 
derived therefrom), ^ind. brekka figures frequently in Icelandic 
local names. 

Seaforth. — This modern residential suburb is an 
instance, with Liverpool and Hoylake, of a township bear- 
ing the name of a sheet of water ; the present appellation 
being borrowed by Mr. Gladstone's father from the Scottish 

1 Hist, of Sankey, 1889, p. i. 

2 Proceedvigs Lit. and Phil. Soc, L'pool, vol. xxi. (1S66-67), p. 64. 


Seaforth, which is derived from the O. Nor. scsr (accus. see), 
' sea,' 2indifiorthr, ' firth,' ' frith,' ' bay.' 

Sephton or Sefton. — Domesday has Sexlone ; the 
Valor of Pope Nicholas (1291) CeJ'ton ; the Testa de Nevill, 

The Rev. Geo. W. Wall, Rector of Sephton, writes to 
me : " Up to the death of a rector in the sixteenth century, 
' Sefton ' obtains in the registers ; but his burial is entered 
as ' of Sephton,' and ' Sephton,' with few exceptions, is the 
ecclesiastical spelling thenceforward, and is engraved on the 
church plate, etc. . . . ' Sephton ' appears the later use. . . . 
I do not think that the Lancashire and Cheshire Historical 
Society . . . have ever attempted a derivation." 

Unless the Sextone of Domesday is a clerical error, it 
might imply that the original farmstead was Seaf's tun. We 
find this name, combined with the patronymic suffix ing, in 
such English village names as Seavington and Sevington. 
On the other hand, the designation of the township under 
discussion occurs so consistently in post-Domesday times 
without a medial genitive s, that it seems impossible to 
reject the conclusion that Sextone is a blunder of a Norman 
scribe, and that the name must be referable to the O. Nor. 
.y6/(Swed. sdf^ Dan.-Norw. siv, A.-Sax. secg), 'sedge.' This 
definition would be accurately borne out by the natural 
characteristics of the district, which is low and swampy. 
We find the word sef in such Icelandic local names as sef- 
dcela, 'sedgy hollow,' sef-tjorn, ' sedge-tarn.' 

Skelmersdale — Uctred, at the time of the Domesday 
Survey, held Schelmeresdele. The literal meaning of the 
name is 'the devil's dale' — O. Nor. skelniir (genit. sing. 
skelniis), ' devil ' ; but Skelniir was doubtless the name of the 
original Scandinavian proprietor. 

Southport received its name in a somewhat arbitrary 
fashion at a dinner which a Mr. Sutton, of North Meols, 
gave in 1792, in celebration of his founding a hotel on the 
site of the now favourite watering-place. 

Southworth. — The latter element of this name is the 
A.-Sax. weort/iig, ' farm,' ' estate.' 


Speke. — Spec is the Domesday rendering. There is only 
one Speke in England. The name may possibly be referred 
to the A. -Sax. spc'cc-hus^ ' a court hall ' — spc'cc, ' speech '; hus^ 
' house,' ' hall ' — hence the famous Speke Hall ■} but it is 
more probable that it is the A.-Sax. spic (Ger. speck), which, 
as Kenible points out, ^ properly signifying 'bacon,' was used 
to denote swine pastures. 

Spellow. — Anciently Spellawe. In the second element 
we have the common A.-Sax. hheto, ' hill,' ' mound ' (as in 
Lotv Hill) ; the first syllable in names formed similarly to 
this (for example, Spelhoein Northants) is generally ascribed 
to the A.-Sax. spell, ' speech,' Spellow thus meaning Speech- 

St. Helen's derives its name from the old episcopal 
chapel of St. Helen — called St. Ellen in 1650. 

Stanley. — The name of this Liverpool suburb is derived 
from the noble family bearing that patronymic, which signi- 
fies ' the stony or rocky meadow ' — A.-Sax. sldn, ' stone,' 
'rock'; ledli, 'meadow,' 'lea' — and is ultimately to be 
referred to Staffordshire. " The family of Stanley is a branch 
of the ancient barons of Audeley or Aldelegh, in Stafford- 
shire, one of whom, Adam, had two sons, Lydulph and 
Adam. The former, Lydulph de Aldelegh (lemp. Stephen), 
was progenitor of the barons Audeley : the second son, 
Adam, assumed the name of Aldithlega or Audleigh, and 
had a son, William, to whom his uncle Lydulph gave Stan- 
leigh or Stoneleigh, in Staffordshire, on which he assumed 
the surname of Stanley."^ See Winstanley. 

Tarbock. — Domesday has Torlwc. The township takes 
its name from the local beck or brook, which was anciently 
known as the Torhec, that is, 'the strenm of (the Scandi- 
navian deity) Tor, or Thor ' — O. Nor. bekkr, Dan. bcek, 
Swed. biick, ' brook.' 

Thingwall. — "Between the parishes of Childwall and 
West Derby, but included in neither of them," say the 

^ In later times the ' speke-house ' was the parlour or reception-room 
of a convent. 

■^ Codex Diplomaticus, iii. xxxvii. 
^ Baines, Lane, 1870, ii. 271. 



editors of Baines's 'Lancashire' (1870, ii. 387), "lies the 
hamlet of Thingwall. Being extra parochial, and now con- 
sisting of a single estate, and without any dwellings on it 
except the mansion of the proprietor, and an old farm or 
manor-house, it seems to have escaped the notice of topo- 
graphers, is unnoticed in the census returns, and received 
no mention whatever in the original edition of this work ; 
yet it gave a surname to an ancient family, and is men- 
tioned as a distinct manor in the early ' extents ' and ' in- 
quisitions ' of the county, while its name affords one of the 
most distinct traces of the early settlement of this part of 
the county by Scandinavian invaders." 

This Thingwall (O. Nor. //zzV/j^, 'parliament,' vd7/r (dat. 
case 7'e//i), ' field ') undoubtedly marks the spot at which the 
Norsemen of south-west Lancashire were accustomed to 
meet in council, promulgate decrees, and transact other 
business of importance, just as the Thingwall in Cheshire 
was the parliament-field of the Wirral Norsemen ; while 
Tynwald is the parliament-field in the Isle of Man to this 
day. The name occurs, more or less disguised, in other 
portions of the country which were settled by the Danes and 
Norsemen. The modern Norwegian sior-t{h)ing is ' the 
great court' or 'parliament.' It seems to have been the 
aim of the Scandinavians in choosing a tliiiigvollr to select 
a plain in the middle of which rose an eminence upon 
which the chief men could take their stand and address the 
people upon the lower levels around them. It is thus that 
in some cases the Norse name thingvollr or ' thingwall ' 
eventually came to denote the eminence alone instead of 
the flat expanse around it. 

The Lancashire Thingwall is not mentioned in Domes- 
day, but the chief manor of Derbei Hundred is recorded in 
that Survey as containing six berewicks, or subordinate 
manors, and it is considered that Thingwall was one of 

Thornton. — The ash, the oak, and the thorn (A.-Sax. 
\orn) have supplied the bulk of English place-names derived 
from plant-life. The various Thorntons are frequently re- 
presented in Domesday, as in this instance, by Toreiitun. 
The Lancashire Thornton occurs in the Testa de Nevill as 


Toxteth. — This name occurs in Domesday as Stoche- 
stede, i.e., 'the stockaded or enclosed place' — A.-Sax. sfocc 
(Ger, stock), 'stake'; A.-Sax. stede (Ger. stadt), 'place,' 
' stead.' 

Tuebrook is the name of a small stream which has 
given a designation to a residential district of Liverpool. 
The late Sir J. A. Picton thought he detected a Celtic word 
meaning ' muddy ' in the first syllable^ ; but it is probable 
that, as ' brook ' is Anglo-Saxon {broc), so also is ' Tue ' — 
doubtless Tiw, the god of war, as in 'Tuesday,' A.-Sax. 
Tkves-diCg. ' Tuebrook ' is therefore a variant of ' Thorburn,' 
where the latter does not represent the Norse heroic per- 
sonal name Thorbiorn {l)wrn = hea.r). 

Tyldesley.— In the printed Testa de Nevill the name 
of this township occurs as Tyldisley, Tyldesley, and Tydesley 
on fol. 396, and as Tydesle thrice on fol. 398^ It is impos- 
sible from the known existing documentary evidence to say 
which spelling is nearest the original form, or whether the 
first / in the present-day form and in some of the earlier 
spellings is or is not an epenthetic intrusion, due merely to 
a rustic assimilative pronunciation, somewhat, perhaps, in 
the way that Gibcl-al-Tarik was ultimately resolved into 
Gibraltar. Were the first / really intrusive, the name would 
probably have an affinity with the Derbyshire Tideswell. 
In any case the first element is most likely a personal 
appellation. We could scarcely connect the A.-Sax. ///5 
(genit. lilies), ' tilth,' ' cultivation,' etc. (from tilian, * to 
till '), with a ley or meadow. 

UphoUand, near Wigan, occurs in Domesday as 
Holland, and there is no doubt that, as in the case of 
Downholland, this name is referable to an original A.-Sax. 
Holt-land, ' wood-land.' Upholland lies much higher than its 
antithesis, Downholland ; hence the later prefix ; and it is 
still contiguous to woods. Sec Downholland. 

Walton. — The Domesday form is Waletonc. There 
are many Waltons in England, and as a rule we have to 
distinguish whether the name was applied to a walled tun 
or settlement (A.-Sax. weall, ' wall ') or to an enclosure 

^ Memorials of Liverpool, i. 4. 



hemmed in by a weald (A.-Sax. weald, Ger. wald) or wood. 
There is, I think, no difficulty here : " In 33 Edward I. 
[1305] William de Waleton impleaded Robert Byroun and 
forty-six defendants for cutting down oak and other trees 
growing in Waleton, under the pretext that the townships of 
Waleton and Kyrkeby were united by a wood in which they 
had the privilege of husbote."^ 

Wargrave. — This name has troubled topographic 
philologists as to both its elements. C. Blackie^ does not 
mention it ; but the untrustworthy Edmunds" glosses it as 
' the ditch enclosure.' War has been variously taken to 
represent E. Eng. werre, 'war,' A.-Sax. 7vcer, 'sea,' A. -Sax. 
7vdr, ' seaweed,' A.-Sax. waroth, ' shore,' A.-Sax. warn, 
' defence,' A.-Sax, ivyrt, ' wort,' O. Nor. ver, or A.-Sax. wer, 
'fishing-station,' O. Nor. vorr, 'landing-place/ and even 
A.-Sax. wer, 'man '; while ^^-az'^ has generally been thought 
to be A.-Sax. graf (grave), presumed to mean ' ditch,' or 
' trench '; but the Fr. greve, ' strand,' has also been men- 
tioned. Study of this and somewhat similar names leads 
me to think that ' Wargrave ' indicates ' the grove by the 
fishing-station' — A.-Sax. rf^r, 'weir,' 'dam,' 'fish-trap,' 'fish- 
pond ' ; A.-Sax. grdf, ' grove.' Ware, on the Lea, is the 
spot where the Danes raised a great weir, or dam. Ware- 
ham (anc. Werham), in Dorsetshire, was defined by Lewis, 
in his 'Topographical Dictionary,' as 'the habitation on the 
fishing shore.' It is, however, not impossible that ' War- 
grave ' may in some instance represent an original A.-Sax. 
wir-grdf, 'myrtle grove.' 

Warrington. — Apart from the Domesday rendering, 
Walintune, which is probably an error, the earliest-found 
spelling of the name of this town is in a document esti- 
mated to be of the end of the twelfth century, which has 
Weri?igton, i.e., the hin or settlement of the Wserings, mem- 
bers of which clan also held estates at Werrington in Devon- 
shire and Northampton, 'Wehringen and Weringhausen in 
Germany, and perhaps at Varengreville in Normandy. 

^ Baines, Lane, 1870, ii. 284. 

2 Diet, of Place-names, 3rd ed., 1887. 

' Names of Places, 2nd ed., 1872, p. 307. 


Waterloo. — Where it is indigenous, this place name 
simply means ' the watery lea or meadow.' In the case of 
the interesting Liverpool suburb it presents no etymological 
significance. The place barely existed in 1835, and ulti- 
mately rose into being round a hostelry called the Waterloo 
Hotel. After Wellington's victory over Napoleon, the name 
Waterloo became very popular, one of the principal bridges 
over the Thames being christened from the Belgian battle- 

Wavertree. — Domesday has Wavretreu ; but the place 
has also been anciently styled Wartre, Waudter, IFave, 
Wavre, and even Wastpull and Wastyete. The editors of 
Baines's 'Lancashire' (1870, ii. 267) were of the opinion 
that "All these forms of the name bear unanimous testi- 
mony to the barren, uncultivated nature of the district — 
Wastyete, or Wastgate, the gate or road over the waste ; 
Wastpull, the pool on the waste ; of both of which there 
are remains to this day in the village green and the pool. 
There is not a doubt that, until within a recent period, a 
great part of Wavertree and the neighbourhood was unen- 
closed and consequently uncultivated. So recently as 1769, 
on Yates's map of the country round Liverpool, Childwall 
Heath stretched from Wavertree to Woolton." 

It is, however, manifestly incorrect to say that "«// these 
forms of the name bear unanimous testimony to the barren, 
uncultivated nature of the district." The terms Wastyete 
and Wastpull can have absolutely nothing to do wuth 
Wartre, Waudter, Wavre, etc., and were evidently either 
simply alternative designations of the district under discus- 
sion, or designations of different portions of it ; while the 
latter group of names are clearly corruptions or contractions 
of the Domesday rendering, Wavretreu, which is well pre- 
served in the modern name. Canon Taylor, in his ' Words 
and Places' (1864, p. 240), classified 'Wavertree' as a 
Welsh name, and this allocation might have seemed to 
receive some confirmation from the fact that a lithic circle 
was discovered here, which has been thought to be Druidic; 
but the common Southern Celtic tre, 'village,' is invariably a 
prefix in accord with the genius of the Celtic tongues ;^ and 

^ " In this township are found traces of very ancient inhabitants in 
the Calder-stones, a small circle of diminutive monoliths on the S.E. 


in his later work, 'Names and their Histories' (1896, 
p. 377), the Canon has wisely transferred the name to the 
simple English ' tree ' class. As to the probable meaning 
of the first element of ' Wavertree,' I cannot, perhaps, do 
better than quote a note which I have received from Pro- 
fessor Skeat. " Chaucer," he says, "has wipple-tree for the 
cornel-tree, meaning 'waving-tree,' and the A.-Sax. wcBfer = 
always on the move, vibrating. And waver-tne would be a 
splendid word for an aspen." 

West Derby. — The mediaeval importance of this place 
(Domesday, Derbei) is well testified by its giving a name 
to one of the six hundreds of Lancashire. The midland 
Derby, from which the Lancashire Derby has been dis- 
tinguished by the addition of the prefix West, occurs in the 
Saxon Chronicle as Deoraby, and it has been usual to con- 
sider both places as being originally ' the location of wild 
animals'— O. Nor. dyr (A.-Sax. dear— wihence Eng. deer — 
Ger. ihier), ' (wild) beast,' combined with the customary 
Scandinavian ' settlement ' or ' village ' sufifix, by{r), a deri- 
vation which is not inconceivable, so far as the Lancashire 
township is concerned, when we call to mind the forestral 
nature of the district in olden times, and recollect that 
(West) Dyrby, or Derby, seems to have been used as a kind 
of centre at which hunting expeditions were organized ; but 
it is, nevertheless, not improbable, considering how many 
English bys have a personal appellation as their prefix, that 
(West) Derby may have been given the name of its founder. 

boundary of the township, and in the sepulchral remains which were 
disinterred in 1867 by the men engaged to excavate the foundations of 
two houses in Victoria Park. These remains consisted of eight cinerary 
urns of coarse red-burnt clay, in which were human ashes. ... A 
well-formed arrow-head, two scrapers, and other tools of flint, were 
found in immediate proximity to the urns. The Calder-stones form a 
circle about six yards in diameter, and consist of six stones, five of 
which are still upright. They are of red sandstone, all different in size 
and .shape."— Baines, Lane, 1870, ii. 267. Sir J. A. Picton refers to 
these stones in the opening chapter of his Memorials of Liverpool ; 
and the Wavertree residence of Sir John T. Brunner, Bart., M.P. — 
' Druids' Cross ' — has so been designated from them. The word 
'calder' is probably the A.-Sax. f;aldor, 'enchantment,' 'divination,' 
'magic,' etc., occurring in such compounds as galdor-leoth, 'magic 
song'; galdor-word, 'magic word,' 'incantation word.' Compare also 
A.-Sax. galdere or galdar, ' wizard,' 'enchanter.' 


As Thorpe points out, in his glossary to 'Beowulf,' the word 
deSr {dyr) applied to a warrior does not, as in modern usage, 
imply reproach, any more than do the names Wulf, Biorn 
(bear), Hengest (stallion), Horsa (horse), etc. 

Whiston. — This name is a corruption of the A.-Sax. 
hivit-stdn, 'white stone.'^ The old Whiston Hall and its 
outbuildings, still to be seen, are built of white stone. The 
same etymology applies to another Whiston — that near 
Rotherham — where are large quarries of white stone. The 
Lancashire Whiston occurs, in fact, in the Nonarum Inquisi- 
tiones {temp. Edw. III.), fol. 40^, as Whitsta?i. 

Widnes. — The spelling in 1285 was Vidnes : other later 
renderings have been Wydnes and Wydness. This is gene- 
rally taken to be ' the wide nose (or promontory) ' — O. Nor. 
vidhr (Dan. and Swed. vid') ' wide' ; and nos or nes (Dan. 
nase, Swed. ndsd) ' nose ' ; but as the Widnes promontory 
does not seem to be particularly wide — at any rate, at the 
present day — it is a moot point whether, instead of the 
O. Nor. vidhr (' wide '), the O. Nor. vidhr (short/), meaning 
' a wood ' — A.-Sax. wudu, Dan. red, Swed. vdd — was not the 
original component. ^ 

Wigan. — The name of this town does not occur in 
Domesday. Wygan is generally the ancient appellation, 
and this has been referred to the A.-Sax. wigan, ' warriors,' 
'soldiers' (sing. 7aiga ; wig, 'war,' 'battle'), tradition 
averring that in this neighbourhood, on the banks of the 
Douglas, King Arthur defeated the Saxons in several san- 
guinary encounters.^ Whether this be so or not, the fact 
remains that large quantities of bones of men and horses 
have from time to time been turned up here. 

Windle. — In Domesday, in the part relating to the land 
between Ribble and Mersey — " Inter Ripam et Mersham" — 

^Analogy: Chaucer's za/ies/on = whetstone — "A wlieston is no 
kerving instrument." — Troil. and Cris., i. 631. 

- Runcorn and Widnes practically form one town, but the former 
does not, geographically, come within the defined scope of the present 
monograph. It can, howe%'er, be stated that the name Runcorn occurs 
in the Anglo-Saxon Chronicle, under the year 913, as RiinicSfa (the 
oblique case is used, ' aet Rumcofan ') — A.-Sax. riiin, 'roomy,' 
'spacious'; cofa, 'cove,' 'creek.' 

^ See quotation from Higden's Polychronicon under Douglas. 


occurs a Wibaldeslci, which would appear to embrace the 
present townships of Windle, Whiston, Bold, Prescot, etc. 
As, however, there is no place existing in Lancashire with a 
name resembling Wibaldeslei, while there is a Wimboldsley 
in Cheshire, I am prepared to go further than Beamont, 
who thinks^ " it is possible that the name is misplaced here 
[Lancashire]," and to assert that the entry of Wihaldeslei as 
being " inter Ripam et Mersham," must, beyond doubt, be 
a mistake on the part of a Norman scribe. As to Windle, 
in the Testa de Nevill we find the spellings Wyndul (fol. 
396) and Wyndel (398^). This may mean ' the pleasant 
dale ' — A. -Sax. 7i>yn{?i)-dcel ;~ or the first syllable may be 
from a personal name. See Winstanley, Winwick. 

Winstanley. — There is but one place of this name in 
Great Britain. It does not occur in Domesday, but in the 
Testa de Nevill (fol. 406) we find JVinsta?ieleg' ; and in the 
Survey of 13 20- 1346, Chetham Soc, vol. Ixxiv., 1868, we 
have Wynsia?ileghe. There is no etymological difficulty as 
to the latter portion of the name— A. -Sax. sidti, ' stone '; 
ledh, 'lea,' ' leigh,' 'ley,' or ' meadowland ' (see Stanley); 
but the ' win ' is a somewhat doubtful quantity. The whole 
name Winstanley of course simply means ' the meadowland 
belonging to W^instan,' which was a common enough proper 
name. Thus, according to Domesday, a Saxon named 
Winestan held our Walton in the time of the Con- 
fessor ; in a charter of yEthelred's (a.d. 996)^ we have 
** Wynstanes ham " (Wynstan's home) ; in one of Eadward's 
{c. A.D. 910),^ " Winstanes stapol " (Winstan's pillar); a 
" Winstan, minister," attests one of Bishop Denewulf's 
charters (a.d. 879-909)''; and so on. But, as we have said, 
it is a little difficult to trace the exact meaning of 7uin or 
wyn here. It certainly cannot be the A.-Sax 7cnn, ' wine '; 
it is scarcely likely to be, in this particular case, A.-Sax. 
witie, 'friend,' or even A. -Sax. 'cvyn or wynn, 'pleasant,' 
' beautiful ' ; and we may reject the AVelsh g7vyn (mutated 
to wyn), 'white,' 'fair' (although the first element of 

^ Domesday Cheshire and Lancashire, 1882, p. 75. 

^ Compare wynn-land, 'the pleasant land,' in the ])oem of the 
Phoenix, ascribed by .Sweet and others to the Northumbrian poet 

* Codex Diplomalicus, vi. 137. ^ Ibid., v. 184. ^ Ibid., v, 163. 


another Saxon name, Dunstan, is probably from the Wei. 
dwn, ' dun,' ' dark ') ; and also the hypothesis of a corrup- 
tion of ' Woden,' as in Winsborough. It is most likely 
A. -Sax. wi?i or wiii?i, 'war.' Bosworth ('A. -Sax. Diet.,' 
1838), quoting Lye ('Diet. Sax. etc.,' Manning's ed., 1772), 
says that from this 7in'n we get the proper names Alwin, ' all 
in war'; Baldwin, 'bold in war '; Eadwin, ' happy in war'; 
Godwin, ' good in war ' ; but it is pretty clear that win or 
7C'ine as a termination in Saxon names is the A. -Sax. zvine, 
'friend,' 'protector' — thus Leofwine = Beloved Friend.^ 
Florence of Worcester, under the year 992, records that 
not long after the death of the blessed Oswald, Duke 
vEthelwine, " friend of God," of distinguished memory, 
passed away.^ As a prefix, however, win seems often to be 
the A.-Sax. win, ivinn, ' war.' IVinstan would therefore 
literally mean 'war-stone ' or 'battle-stone,' which may have 
denoted a monument ; or it may have been an archaic term 
for a battle-axe, or perhaps have indicated a war-(stone) 
house, or castle. This literal rendering of ' battle-stone ' is 
apparently confirmed by the occurrence of the personal 
name Wigstan, which has the same meaning ; although 
Miss Yonge" (who does not give Winstan) makes all her 
win nominal prefixes represent ' friend ' only. 

Winwick. — An old spelling is Wymoyc. This is pro- 
bably ' the pleasant habitation ' — A.-Sax. wy7i ' pleasant ' ; 
7C'/V, 'dwelling,' 'habitation,' 'village,' etc.: wynland, 'the 
pleasant land,' ' the land of joy,' occurs in the poem of the 
Phoenix, attributed by Sweet and others to Cynewulf. 
Compare the word 'winsome.' 

Woolston. — In mediaeval documents the spelling is 
often Wolston, which form occurs in the Nonarum Inquisi- 
tiones, temp. Edw. III., fol. 40, The name must be a 
corruption of A.-Sax. Wulfes-tun — ' Wulf's farmstead.' 

^ See Kemble's Names, Surnames, and Nic-names of the Anglo- 
Saxons, Proceedings Archaiol. Inst. Gt. Eritain, 1846, p. 87 ; and Sir 
J. A. Ficton's Proper Names in Philological and Ethnological Enquiries, 
Proceedings Lit. and Phil. Soc. L'pool, vol. xx. (1865-66), p. 1S8. 

- " Nee diu post excessum beati Oswaldi, egregice dux memorias 
/Ethelwinus, Dei Aniictis, defunctus est." — Elorentius Wigorniensis, 
Thorpe's ed., 1848, i. 149. 

' History of Christian Names, 1884. 


V7oolton. — This name is literally a wolf in sheep's 
clothing : it has nothing in common with the hair of the 
well-known animal any more than that of Woolwich has. 
Little Woolton is identified with the Domesday Ulventune^ 
and Much Woolton with the Domesday Uvetone. Local 
historians have presumed from these names that the district 
at the time of the Saxon (or Norse) settlement was infested 
by wolves (A. -Sax. tvulf, O. Nor. ulfr^ Dan.-Norw. ulv, 
Swed. itlf^ ' wolf) ; but there is little doubt that Ulvejitune 
and Uvetone were originally the Wins or farmsteads of a 
proprietor bearing the common Teutonic name Wulf or 
Ulf. Wolvesey, a little island near Winchester, was, 
however, the scene of the annual payment of the Welsh 
tribute of wolves' heads. 


Arrowe (fourteenth century, Arwe), like Landican, its 
near neighbour, may with comparative safety be deemed a 
relic of Celtic nomenclature — Wei. ern>, ' acre,' ' piece of 
ploughland '; connected with Lat. aro, Gr. dpow, A.-Sax. 
erian, ' to plough '; and hence Eng. * arable.' 

Backford. — The first element is DanoSaxon for 
' brook ' — O. Nor. bekkr, Dan.-Norw. biek, Swed. biick, 
Nor. Eng. beck, Ger. bach. 

Barnston. — The Domesday spelling is Bernestone ; 
while in early charters the name also occurs as Bermston, 
and Bernstotie, as well as Barnston, the present day render- 
ing. It may possibly be due to a combination of a common 
Norse personal appellation meaning ' a bear,' viz., Biorn 
(genit. sing. Biarnar), with the O. Nor. stemn, 'a stone' 
(monument) ; but it is much more probable that the place 
was originally the iun or farmstead of a Saxon settler named 
Beorn (genit. sing, or possess. Beornes), an appellation 
denoting ' one bold in war.' 

Bebington. — This was a branch iun or settlement of 
the Saxon Bobbing family.^ 

Bidston. — The name of this place is found in 1272 as 
Byddeston. There is little doubt that the first element here 
represents the patronymic borne by the Saxon owner 
of the original tun or farmstead, although I place on record 
the suggestion of local archaeologists that, as a runic stone 
has been found at Upton (about two miles from Bidston) 
containing the A. -Sax. verb biddan, ' to pray,' the name of 
1 Birch's ed. (1876) of Kemble's Saxons in England, i. 457. 


the township is an attenuated form of biddende-stdm, ' pray- 

Birkenhead. — Ormerod's " head of the River Birken "^ 
having been satisfactorily disposed of by local antiquaries, 
it is easy to fall back upon the obvious — birken-head, ' head 
or promontory of the birches' — O. Nor. biork, A.-Sax. birce, 
' birch'; O. Nor. hofud, A. -Sax. hcd/od, 'head.' Hence the 
name Woodside. Birkenhead is not mentioned in Domesday. 
The earliest recorded spellings of the name are : Birkefi- 
heved, jBirkhe/ied, Byrkehed, Birchened, Byrchened, Byrken- 
hed, Byrkehe?ied, Birkynhede, etc. 

Blacon. — The Domesday spelling is Blachehol, which in 
later times seems to have been corrupted into Blaken. 
With the Norman ch as usual equating k^ Blachehol clearly 
represents ' black hole ' — A.-Sax. blac, ' black,' and hoi, 
'hole,' 'hollow.' A connection with the A.-Sax. bide 
(i) 'bright, (2) 'bleak,' would scarcely be appropriate. 

Brimstage. — This name occurs in early documents as 
Brimstath and Brynstath (and even Bryfisfon), indicating 
that this was the stead or place — O. Nor. stathr (A.-Sax. 
stede) — settled by a Norseman named Brun or Bryn — O. Nor. 
brihin (A.-Sax. brihi, Scand. brun), ' brown,' ' dark.' This 
personal name frequently occurs in early charters. See Bryn 
and Bromborough. 

Bromborough. — Before attempting to deal with the 
etymology of this name, it is necessary to consider the 
evidence for and against the identification of Bromborough 
with the Brunanburh around {ymbe) which ^^ithelstan, in 
A.D. 937, achieved his great victory over the allied Danes, 
Irish, Scots, and Welsh. The site of the battle of Brun- 
anburh has long been a subject of controversy, but until 
comparatively recently the claims of Bromborough to be 
considered the scene of the sanguinary conflict, probably 
owing to the former secludedness and insignificance of the 
township, have scarcely been thought worth discussing. 
Thus Gibson merely mentioned the fact that there was a 
place in Cheshire called ' Brunburh,'- a statement which 

' Chesh., 1819, ii. 254. 

^ " Oppidum est in agro Cestrensi hodie Brunburh dictum." — 
Chronicon Saxoiiicum, 1692. 


Bosworth ('A.-Sax. Diet.,' 1838) repeats. Thorpe, in his 
edition of the Saxon Chronicle (1861), was unable to locate 
Brunanhurh ; so was Earle in his {1865); but Plummer, 
re-editing Earle's edition in 1889, queries the county of 
Durham, as advocated in Bosworth and Toller's ' A.-Sax. 
Diet.' (1882), and prefers, with Powell, to think that the 
battle was fought in Lancashire. Thomas Baines, however, 
in 'Lancashire and Cheshire, Past and Present,' 1867, 
i. 316, was of the opinion that it took place near Brom- 
borough in Wirral. 

Some correspondence on the subject is to be found in 
the Athencuwi of the second half of 1S85. In the issue of 
that journal for August 15, 1885, p. 207, Dr. R. F. Wey- 
mouth entertains no doubt that Bromborough in Cheshire 
is Brunanburh, and he speaks of " traces of a great battle in 
that neighbourhood." In the issue for August 22, 1885, 
p. 239, the Rev. T. Cann-Hughes points out that the ques- 
tion has been discussed in the Cheshire Sheaf, that Mr. 
John Layfield shows that on the Ordnance Survey for 
Bromborough parish the ' Wargreaves ' is mentioned as the 
site of a battle between ^thelstan and the Danes in 937, 
and that in the Proceedi?igs of the Chester Archceological 
Society (vol. ii.) there is a paper by the secretary, Mr. 
Thomas Hughes, in which it is stated that about 910 the 
Princess ^thelfleda built a fortress at Brimsbury, which is 
identified by local authorities with Bromborough. Another 
contributor to this correspondence, however, asserts Brunan- 
burh to be in Dumfries-shire ; another claims it to be near 
Axminster, while Mr. Herbert Murphy, writing in the 
Athenmwi of October 3, 1885, p. 436, thinks that Mr. 
Hardwick, in his 'Ancient Battlefields in Lancashire' 
(1882), has made out an irresistible case in favour of the 
country round Bamber Bridge, just south of Preston and 
the Ribble, stress being rightly laid on the discovery, in 
1840, in this locaHty of the famous Cuerdale collection of 

On the other hand, Dr. Birch, in his ' Cartularium Sax- 
onicum' (1885, etc.), ii. viii., maintains that Brunan-burh is 
a poetical alliteration for Brinnnga feld, which occurs in a 
Latin charter of King ^thelstan, a.d. 938 (' Cart. Sax.,' 
ii. 435), and, arguing that an English Broomfield or Brom- 


field must supply the site of the conflict, he suggests Broom- 
field in Somersetshire. I must confess that this portion of 
Dr. Birch's reasoning does not convince me. Brunnan- 
burh or Brunan-burh may be a form in which historical 
accuracy is sacrificed to poetical demands ; but the fact 
that a charter refers casually to the battle having been 
fought at or in Bruninga-feld need not count for much. 
This name strikes one as a generalization, meaning simply 
'the plain of the Brunings,' i.e., of the descendants of 
Brun ; and, in fact, this occurrence of Bruninga-feld might 
seem to some to tend to the confirmation of the theory that 
Cheshire witnessed the battle of Brunanburh, for in this 
county we have, in comparative contiguity, at least three 
places which may owe their name to an eponymic Brun — 
namely, Bromborough (formerly Brunborough, Brunbree, 
etc.), Brimstage (formerly Brunstath), and Brinnington. 
Besides, as to Bruninga-feld representing a modern Brom- 
field or Broomfield (Bartholomew's Gazetteer gives three 
Bromfields and five Broomfields in England), it must not 
be overlooked that a sharp labial, as/ is, is not so liable to 
convert a preceding n into m as a flat labial like /^ is ; 
and a Bromfield or Broomfield, just the same as a Bromley 
or Brompton, may generally be taken to imply a place which 
was overrun with broom. 

In the map entitled ' Die Britischen Inseln bis auf 
Wilhelm den Erobcrer, 1066,' in the Spruner-Menke 'Hand- 
Atlas fiir die Geschichte des Mittelalters ' (1880), Brunan- 
burh is placed on the ' Meresige ' in about the present 
position of Bromborough. A gentleman who has given 
much study to the question on the spot, the Rev. E. D. 
Green, Rector of Bromborough, wrote to Mr. Helsby, the 
editor of Ormerod ('Hist. Chesh.,' 1882, ii. 427): "A 
large tract of land near the seashore at Bromborough has 
long been known by the name of IVargraves. This fact, 
and that of the recent discovery (June, 1877) of a large 
number of skeletons near the coast of the Dee, a few miles 
further off, with other circumstances, combine to prove that 
this parish was the unquestionable^site of ^thelstan's famous 
victory over the Danes and their allies in 937." 

There are one or two other points which would appear 
to add strength to the theory of the Battle of Brunanburh 


having been fought in Cheshire. In the first place it is 
probable — given the actual existence of a Brunanburh — 
that there was but one Brunanburh in England in a.d. 937, 
just as there is but one Bromborough to-day. Secondly, 
the Dee and the Mersey, whose estuaries are divided by 
the Wirral Peninsula, have, from time immemorial, been the 
favourite points of embarkation for and debarkation from 
Ireland ; it is, indeed, tolerably certain that the first Irish 
missionaries to visit England landed in Wirral.^ Thirdly, 
we know that a considerable Norse and Danish population 
had already settled in Wirral when Anlaf's ships crossed the 
Irish sea, and the Hiberno-Danish king could surely reckon 
upon the support of his fellow-countrymen. 

The fact that certain land at Bromborough is known as 
the Wargraves is, however, of no significance. The Early 
English werre^ ' war ' (if that be the word intended), was 
not in use at the time of the battle, 7vig being the ordinary 
A.-Sax. word, and the one used in the poem-chronicle^ 
itself. The A.-Sax. grcrf (pi. gncfas) — whence Mod. Eng. 
'grave' — certainly meant 'trench,' 'ditch,' or 'pit'; but 
without evidence of early spelling it is not safe to say what 
the war in Wargraves positively represents. (Is it the 
A.-Sax. warn, 'defence,' or A.-Sax. wcer^ 'sea,' or A.-Sax. 
waroth, 'shore,' or A.-Sax. wer, 'fishing-place'?) Besides, 
the terminations ^/-az'tf ^wdi graves in place-names are usually 
attributable to A.-Sax. graf (pi. gnifas), ' grove.' But see 
Wargrave in the West Derby Section. 

The question may now be asked. Is the available evidence 
fairly conclusive in favour of Bromborough being Brunan- 
burh ? I am afraid that the answer must be that it is not. 
And for this reason, namely, that the indefatigable re- 
searches of Mr. T. T. Wilkinson'^ and Mr. Chas. Hardwick,-* 
combined with the Cuerdale find of coins, leave scarcely 
room for doubt that the great battle of a.d. 937 was fought 
in the northern portion of that principal part of Lancashire 

^ See Helsby's ed. Oimerod's Hist. Chesh., 1882, ii. 486. 

- The reader scarcely needs to be reminded of Tennyson's metrical 
version of the Battle of Brunanburh. 

^ Transactious I list. Soc. Lane, and Chesh., 1S56-57, pp. 21 sqij. 

■• Hist, of Preston, 1857 ; and Ancient Batllcliekis in Lancashire, 


which Hes between Ribble and Mersey. Mr. Wilkinson 
makes out a very good case for the neighbourhood of 
Burnley, which is on the river Brun,^ and was formerly 
known as Brunley, while close by are Saxifield and Danes 
House. Mr. Hardwick argues for the district south of 
Preston, and points to such names as Bamber and Brindle. 
A theory reconciling these two diverse views would make 
out that the battle was actually fought at or near Burnley, 
that the defeated Danes and Irish were pursued to their 
ships in the Ribble, and that, when that river was reached, 
the chest constituting the Cuerdale find had to be hurriedly 
buried to prevent its falling into the hands of the enemy. 
The Cuerdale treasure-trove, it may be recalled, consisted 
of (besides ingots, etc.) some 10,000 silver coins enclosed 
in a chest. The greater number of the coins were Danish ; 
a large number were Anglo-Saxon, and a smaller number 
were French, the remainder being made up of Italian and 
Oriental pieces. ^ The fact that specially interests us now, 
however, is that a/l these coins ivere minted between a.d. 815 
and A.D. 930, and they must consequently have been in- 
humed within a comparatively short period after the latter 
date, that is to say, about the time of the Battle of Brunan- 
burh. Worsae remarks,^ with needless caution, that "the 
treasure must have been buried in the first half of the tenth 

As we have therefore decided that Bromborough is not 
Brunanburh, it will be as well to note the early forms of 
the name Bromborough as they are given by Mr. Green, 
who writes -^ " In 912 we have it ' Brimburgh,' and before 
the Conquest it is ' Brunsburg,' ' Brunnesburgh,' and 
' Brimesburgh ' ; in 1152 ' Brunborough ' ; tern. Pope 
Honorius, ' Brumbure ' ; tern. Edward I. (in its charter) 
' Brumburgh ' and ' Bromburgh ' ; in 1291 ' Bromborch ' ; 
in 1548 ' Brombrogh,' 'Brumburgh,' and ' Brumborowe ' ; 
tern. Eliz. ' Brumbrow ' ; in 17 19 ' Brombrough,' and since 

^ It should be noted that in the Annales Cambria; we have "Bellum 
Brune," and in the Brut y Tywysogion, or Chronicle of the (Welsh) 
Princes, "Ac y bu ryfel Brun"— " And the Battle of Brun took place." 

' See the Ntimismatic Chronicle (Journal of the Numismatic Soc. ), 
1842-43, pp. 1-120, with ten plates. 

' Danes and Norwegians in England, 1852, p. 49. 

^ Helsby's Ormerod's Hist. Chesh., 1882, ii. 427. 


' Bromborough ' (pron. Brumbo rough)." The pre-Conquest 
forms point to the personal name Erun^ — A. -Sax. brihi, or 
O. Nor. hriiim, 'brown,' 'dark' — combined with A.-Sax. 
biirh or burg, or (J. Nor. borg, 'castle,' 'fortress';- and I 
am therefore unable to agree with Mr. Irvine's derivation of 
the first element of 'Bromborough,'^ viz., O. Nor. brunnr, 
' well,' ' spring.' See Brimstage (Wirral Hundred) and 
Bryn (West Derby Hundred). 

Burton. — This is a common English place-name, which 
Taylor says (' Names and their Histories,' p. 79) "is usually 
from the A.-Sax. bi'/r-ti'oi, which denoted a f/hi or farmyard 
containing a bur or ' bower,' the word bur meaning a ' store- 
house ' in O. Nor., and in A.-Sax. a 'chamber,' 'sleeping 
place,' or building of some kind " ; but, as I have pointed 
out to Canon Taylor (who writes that he cannot affirm that 
my point is wrong), this explanation of the name seems 
lame on the face of it, for surely every farmstead, ancient 
or modern, must of necessity have possessed or possesses 
accommodation of this kind. The difficulty is got over, I 
think, by taking the bur in the bur-tuns to represent not an 
apartment but a husbandman, tiller of the soil, working 
farmer, corresponding to the Du. boer, Ger. bauer, Platt- 
Ueutsch buur., from the first of which our lexicographers, 
including Skeat, have derived the Eng. 'boor,' whereas 
' boor ' has a continuous history in England, from A.-vSax. 
iyge) bur onwards through Middle English, and it is found 
in ' neighbour,' originally ' the near farmer.' See Bosworth- 
ToUer's 'A.-Sax. Diet.,' and Stratmann-Bradley's ' M. E. 
Diet.', s.v. The bur seems to have been near akin to the 
ccorl. The synonymy of mod. boor and churl apparently 
proves this. Taylor enumerates some 60 Burtons and 
77 Carltons, Charltons, and Chorltons — not a very wide 

^ Brunswick in Germany, formerly Brtines7uic, represents ' Bruno's 

- It is curious to note that, while the A.-Sax. Inirg, originally 
signifying an isolated stronghold (from A.-Sax. beorgan, 'to protect'), 
acquired the signification of 'city' at a very early period, the Scand. 
/'^/y still largely tends to retain its primary meaning of 'castle,' 'fort,' 
'palace.' This is, of course, due to the vastly different economic 
conditions which have prevailed in England and Scandinavia. 

3 Transactions Hist. Soc. Lane, and Chesh., 1S91-92, pp. 279 sqq. 



In the Laws of Ine (a king of Wessex), sec. ' Be gefeoh- 
tum ' (Fights), it is interesting to note how the scale of 
punishment fluctuates according as the brawl takes place 
in the king's palace, an abbey, a nobleman's or high official's 
residence, or the homes of the taxpayer and the 1>ook 

Sir Henry Ellis points out^ that Lord Coke calls the 
Bordarii of Domesday ^'/^oors holding a little house with 
some land of husbandry, bigger than a cottage " ; and Sir 
Henry further observes :- " The Bures, Buri, or Burs are 
noticed in the first volume of Domesday as synonymous 
with Coliberti. . . . The name of the Coliberti was un- 
questionably derived from the Roman Civil Law. They are 
described by Lord Coke as tenants in free socage by free rent." 

Oaldy. — Domesday has Calders, which may refer to the 
two Caldys — Great and Little. Caldy is Scandinavian for 
' bleak island ' — O. Nor. kald{r) (Dan. kold^ Swed. kall)^ 
' bleak, ' ' cold ' ; ey, ' island ' ; — but as Caldy can scarcely 
be called an island — at any rate at the present day — this is 
probably an instance where O. Nor. ej>, as is sometimes the 
case with A. -Sax. ?§■ ( = iy), represents a place beside water 
rather than land entirely surrounded by water. 

Capenhurst occurs in Domesday as Capeles, which is 
Old l'"rench (Low Lat. capella) for ' chapels '; but ' Capen- 
hurst ' makes its appearance early in the fourteenth century. 
The ' hurst ' (A. -Sax. hyrst, ' wood ') is, seemingly, a post- 
Domesday suffix, and the present name, although Mr. Irvine^ 
thinks it may represent " the brushwood where the capon 
was reared," would appear to be simply a corruption of, or 
false analogy for, Capelhurst (or Cape/shurst), ' the wood by 
the chapel (or chapelsj.' "The township," wrote Ormerod,* 
" is judiciously broken by plantations." 

Chester was the Deva of the Romans, from the name of 

1 Introduction to Domesday, 1816, xxvi. 

- Ibid., xxvii. See also Birch's ed. (1876) of Kemble's Saxons in 
England, i. 131, 215, 216, 225, 226 ; and Seebohm, English Village 
Community, 1884, pp. 131 sqq. Prof. Maitland says: — " Next above 
the sa-vi we see the small but interesting class of burs." — Domesday 
Book and Beyond, 1897, p. 36. 

•* Transactions Hist. Soc. Lane, and Chesh., 1891-92, pp. 2']() sqq. 

* Chesh., 1819, ii. 314. 


its river (see Dee).^ It was long the headquarters of the 
famous Twentieth Legion, and became known as Castrum 
Legionis, or Castra Legionis, ' fort ' or ' camp of the legion,' 
and Civitas Legionuiii, ' city of the legions.' The Britons 
called it Caer Llcon — Wei. caer, ' fort,' ' city '; Ikon, formed 
from legione, the ablative case of Lat. legto, ' legion.' The 
Saxons in their turn styled it Legaceaster, 'the legion Chester' 
(A.-Sax. ceasfer, ' city,' from Lat. castra, ' camp '). In the 
Saxon Chronicle, in the portion dealing with Alfred's wars 
with the Danes, we read (a.d. 894) that the latter "gedydon 
on anre westre ceastre on Wfrhealum seo is Legaceaster 
gehaten " (arrived at a waste Chester in Wirral which is 
called Legaceaster)." The prefix was eventually dropped. 

Childer Thornton. — This place is not mentioned in 
Domesday. The name has suffered no alteration in the 
course of centuries. Saxon Thorntons (tuns or farmsteads 
which took their name from the thorn-bush) are common 
enough in England, but this is the only one with the prefix 
C/ii/der, which is the Mdle. Eng. nominative and genitive 
plural of cliild-e (A.-Sax. cild). ' Childe,' in the Middle 
Ages, was a title of honour borne by the sons of noblemen — 
as a rule until they attained the rank of knighthood. The 
term is familiar to readers of the present day through Byron's 
' Childe Harold.' In local names it seems to have survived 
in Child's Wickham (Glouc), Child's Hill (M'sex.), Child's 
Ercall (Salop), Childerley (Cambs.), etc. Bardsley, in his 
' Enghsh Surnames ' (pp. 202, 534), quotes from mediaeval 
records such names as Ralph le Child, Walter le Child, 
Roger le Childe, etc. It is not probable that the above-men- 
tioned place-names have anyconnection (as somehave hinted) 
with the O. Nor. kelda, Dan.-Norw. kildc, 'well,' 'spring.' 

Cliorlton. — The Chorltons in England are not so 
numerous as the paronymous Charltons and (Nor. Eng.) 

1 "How Deva came to be the name of Chester or the Castra 
Legionis (whence the Welsh Caer Lleoti) is not clear ; possibly it was 
at first the camp Ad Devam, or ' by the Dee.' " — Rhys, Celtic Britain, 
1884, p. 292. 

^ " Civitatem Legionitm, tunc temporis desertam, quae Saxonice 
Legeceaster dicitur . . . intrant." — Florence of Worcester, a.d. 894. 
And " Civitas quK Karle-^ion Britannice. et Legeceastre dicitur Saxonice." 
—tlbuL, A.D. 908. 


Carltons — A. -Sax. ceor/ (Eng. c/mr/), Scand. kar/, ' common 
man,' ' fellow.' The Carltons, Chailtons, and Chorltons, 
like the Hintons, were originally no doubt, as a rule, the 
living-places allotted to the agricultural and pastoral helpers 
on a large estate, but it is probable that some were little fihis 
or farmsteads cultivated independently by small freemen. 

Claughton was formerly called Claghton, Claightoti, and 
Claytofi, names evidently due to the large beds of clay 
(A.-Sax. dc'eg) found in the township. 

Croughton or Croghton.— It is somewhat difficult to 
give with certainty the meaning of the first element of this 
rather uncommon Saxon place-name. Did the name, as 
appears most likely, signify originally ' the crook or crooked 
ti'ni,' or rather ' the iiin in the crook, bend, or corner ' — 
Mdle. Eng. crok, croc, O. Nor. krokr (Dan.-Norw. krog, 
Swed. krok), ' anything crooked,' ' nook ' or ' corner '? Or 
did saffron (A.-Sax. croh) perhaps grow there? Or was 
crockery (A.-Sax. croc, crog, crogh) made there? See Crox- 
TETH in the West Derby section. Croughton is the Domes- 
day Crostone, the s, as in Lesto7ie = Leighton, being the 
Norman representation of the Saxon guttural. 

Dee. — It has been usual to consider that the Celtic name 
of this river, and of others similarly designated, represented 
'dark (water)' — Wei. d/i (dee), Gael. d/i/>/i (duv, doo), ' black,' 
' dark ' (see Douglas in the West Derby section), just as 
the Don rivers were considered referable either to Gael. 
doimJine, * deep,' or donn, 'dark brown ';^ but Professor Rhys 
has a more elaborate explanation. The word Deva, he says,^ 
"originally denoted 'the river,' or rather 'the goddess of 
the river,' for Deva is only the feminine corresponding to 
a masculine devo-s, ' a god ' ; but when the old terminations 
were dropped dcvos and deva assumed the same form, and 
this, according to rule, yielded in Old Welsh doiu or duiu" 
The modern Welsh name of the Dee is Afo7i [river] Dyfrdivy. 

^ See a paper, ' Place-names of Scotland,' by the late Prof. Blackie, 
in Blackwood for July, 1894, in which the following works are passed 
in review: (1) Scottish Land Names, by Sir Herbert Maxwell, 1894; 
(2) Place-names of Scotland, by Rev. Jas. B. Johnston, 1892 ; (3) 
Place-names of Argyllshire, by Prof. Mackinnon, in the Scotsman, 1888 ; 
(4) Place-names of Strathbogie, by Jas. MacDonald, 1891. 

^ Celtic Britain, 1884, p. 291. 


Eastham. — The Domesday spelling of the name of 
Nathaniel Hawthorne's " finest old English village I have 
seen " is Estham. It has been suggested with probability 
that this Anglo-Saxon havim} or extent of land, was so 
named because it lay in an easterly direction from the 
settlement — Willaston — whence the Norman name of the 
Hundred of Wirral was derived. 

Egremont. — This place, like its neighbour. New 
Brighton, is modern, and it takes its name indirectly from 
the ancient Cumberland Egremont, a name which originally 
denoted the castle built on an artificial mount by a Norman 
grantee soon after the Conquest. Egremont signifies ' the 
bold mount ' — O. Fr. egre (Lat. acer), ' bold ' ; nwnt, 
' mount.' Wordsworth has a poem entitled ' The Horn 
of Egremont Castle.' The Cumberland Egremont is referred 
to in a Latin charter temp. King John as Acrimons? 

Ellesmere. — Ellesmere Port is situated at the mouth 
of the canal which commences at the Ellesmere in Shrop- 
shire whence our little port borrowed its name. In ' Elles- 
mere ' we have the same personal appellation that figures 
in the Derbyshire ' Alsop,' which in t)omesday was written 
EJlesJwpe. As in the case of Liverpool and Hoylake, the 
Salopian town has taken the name of the sheet of water 
upon whose shores it has risen into being. 

Fender. — This is the old name of two streams in 
Wirral, upon the northernmore of which the Ordnance 
Survey, for some reason, has bestowed the new appellation 
of Birket. The meaning of ' Fender ' here is rather obscure. 
Some have gone so far as to connect the latter element of 
the name with the Wei. divr, ' water ' ; others observe in 
the former element the A. -Sax. and O. Nor. fen, ' mud.' 
Seeing, however, that the Wirral farmers seem to call any 
kind of large ditch or drain a fender, it is not improbable 
that the term is of comparatively modern origin, and was 
primarily used for the reason that the streams were fenders, 
i.e.^ protectors, against inundations, which were formerly 

^ A. -Sax. ha//i{i>i), 'enclosure,' is distinct from A. -Sax. hain, 'home,' 
which is generally affixed to a personal name. See Leo, Die Angel- 
sachsischen Ortsnamen (Rectitudines Singularum Per.sonarum), llalle, 
1842, pp. 27 sqq. ; and Kemble, Codex Diplomaticus, iii. xxvii., xxviii. 

* Jefterson, Hist, and Antiquities of Cumberland, ii. (1842) 24. 


much more prevalent in Wirral than at the present day. 
Shiices are sometimes called fenders in England (see the 
' Hist. Eng. Diet.,' s.v. ' Fender '). 

Frankby. — This Norse name, omitted in Domesday, 
may well be supposed to indicate a h' or settlement of a 
Frank or Franks. See a footnote under Gayton. 

Gayton. — This place occurs in Domesday as Gaitone. 
I was at first, despite the characteristic Anglo-Saxon termi- 
nation of the name, inclined to think that the tun or 
settlement, like many of the contiguous townships, was of 
Norse foundation, and that the first element of the name 
was referable (i) to the O. Nor. gata (A.-Sax. geaf,a.s in Gates- 
head,^ Highgate, etc.), 'thoroughfare,' 'way,' 'road,' in 
allusion to the fact that Gayton is situated on the direct 
road, on the western or Norse side, between the northern 
and southern ends of the Wirral Peninsula, i.e., between 
West Kirkby and Shotwick ; or perhaps (2) to the O. Nor. 
gei'f" (A.-Sax. gat), ' goat,' which occasionally occurs in 
place-names ; and, again, I judged it not impossible that 
the hamlet might have been founded by a Gedt or Gaut{r), 
i.e., Goth, from Southern Sweden, just as Frankby may 
owe its name to a Frankish immigrant or immigrants.^ 

^ A curious history attaches to the name Gateshead. Bede, by what 
we must consider to be a remarkable error, translated Gdiesheved by 
Caput Caprcv (Goat's Head), although the classical genitive singular of 
A. -.Sax. gat, 'goat,' is gate ox gcete, not gates ; and the error has been 
perpetuated through .Somner (Diet. Sax.-Lat.-Angl., 1659), Bosworth 
(A.-S. Diet., 1838), and even Toller (ed. Bosworth's A.-S. Diet., 
1882, etc.), and Canon Taylor (Names and their Hist., 1896) down to 
the present day. The name is, there seems little doubt (des])ite the 
analogy of other "animal-head" names), referable to the A.-Sax. grat, 
gat, or gat (genit. sing, gates), (i) 'gate,' 'door'; (2) 'street,' 'road,' 
' passage,' and means ' the head or end of the road.' Mr. Richardson, 
Librarian to the Literary and Philosophical Society of Newcastle-on- 
Tyne, writes to me : " I think your explanation of the name of 
Gateshead is the correct one. . . . The road of which it was the head 
is the Roman road from Chester-le-.Slreet, which terminated at Gates- 

^ O. Nor. ei = O. Eng. d, not as pronounced in Southern England, 
like a in ' father,' but as pronounced in Northern England and Scot- 
land, i.e., the Scandinavian-infected jiarts, like a in 'gate.' 

•' In Saxon poems of i:)resumed Scandinavian or Anglian origin 
Swedes and Goths, and also Franks and Frisians, .are mentioned 
together in a similar manner to the Picts and Scots. Thus in "that 


On the whole, however, it would appear that Gayton was 
merely a Saxon goat-farm. We have the analogy of Gatton, 
in Surrey, which is mentioned in the Will of Duke /T'^lfred 
(a.d. 871-889)^ as Gataiun — A.-'&?iX. gdia, genit. pi. of gdl, 
' goat.' A gdta-hus was a goat-house. 

GrOwy. — The name of the river at the southern end of 
the Wirral peninsula is clearly connected with the Welsh 
g7C'y, ' water. '- 

GrGasby. — This Norse-looking name seems really to be 
of Saxon origin, for it occurs in Domesday as Gravcsherie. 
The A. -Sax. grcef^ (O. Nor. grof), the genitive singular of 
which is grfffes, meant (i) 'trench,' 'ditch'; (2) 'grave'; 
while A. -Sax. grdf (genit. sing, grdfes) signified ' grove.' 
Gravesberie, and therefore Greasby, must, however, simply be 
the A.-Sax. Grcefes-burh, * the castle of Gr^ef.' Such pa- 
tronymics as Graves, Groves, and Trench are common 
enough to-day. 

Hargrave. — Domesday has Haregrave. The name 
might seem to indicate that considerable earthworks were 
here thrown up by the usurping Scandinavians. The 
O. Nor. her-grdf (Dan. kcpr-grav, Swed. hdr-graf) may be 
construed as ' military trench '; other similar O. Nor. com- 
pounds include her-floti, ' war-fleet '; her-fblk, ' war-people '; 
her-klcedhi, 'war-clothing,' i.e., 'armour,' the O.Y^ox. herr 
(A.-Sax. here., Ger. heer) primarily signifying 'host,' 'army.' 
As to the easy phonetic interchange of her and har compare 
the various Hardwicks (cattle stations) in England, the first 
element of the name usually being the A.-Sax. heard, ' herd'; 
and Hertford = Hartford or Harford, Derby = Darby, Berk- 
ancient and curious nomenclature of persons and places," The Scop or 
Gleeman's Tale, we read "mid Sweom and mid Geatum " — with the 
Swedes and Goths; while in Beowulf occurs "under Froncum and 
Frysum " — amoncj the Franks and Frisians ; and in The ScAp' or 
Gleeman's Tale, attain, " mid Froncum . . . and mid Frysum " — with 
the Franks and Frisians. 

^ Kemble, Codex Diplomaticus, ii. 120 ; Thorpe, Diplomatarium 
Anglicum, p. 480 ; Birch, Cartularium Saxonicum, ii. igt;. 

^ Prof. Kuno INIeyer (The Voyage of Bran : an Old Irish Saga, 
1896-7, i. 38) says that perhaps the obscure Ir. //a is cognate with 
Wei. s^c'V, and means 'water.' 

^ At the end of syllables and between two vowels A.-Sax. /= 7j. 


shire = Barkshire, etc.^ Again, the prefix har in place- 
names in Danish-settled parts of England may sometimes 
be the O. Nor. hdr,^ ' high '; and ijerhaps O. Nor. /larr 
(A.-Sax. //dr), 'hoary,' 'grey.' (See Greasby.) But in the 
case of Hargrave we think we must agree with Canon Bard- 
sley,^ that it represents simply ' the hare grove ' — A.-Sax. 
Mra, Scand. /mre, ' hare '; A.-Sax. grdf, ' grove ' — just as 
Congreve means ' the coney (rabbit) grove.' 

Heswall. — The name of this township occurs in Domes- 
day as Eswelle ; but in the thirteenth century we have 
HaselweU, and Ormerod, in 1819, calls it Haselwall. The 
name doubtless originally signified ' the field or plain of the 
hazels ' — O. Nor. hasl, 'hazel,' plus the usual O. Nor. v'dllr 
(dat. sing, velli), ' field.' 

Hilbre (Island). — It is really superfluous to add ' Island ' 
after Hilbre, which contains the old word for ' island ' in 
itself. The ancient name was Hildhurgheye^ i.e., ' the isle of 
Hildburgha ' — O. Nor. ey, ' island.'^ 

Hoose. — The name of this little extra-parochial town- 
ship is derived from the plural of A.-Sax. ho (hoo) = ' hough ' 
or ' heel ' (of land). See Hooton. 

Hooton. — Domesday spelling, HotoJU, ' the tihi or farm- 
stead on the hough or heel of land ' — A.-Sax. ho, ' heel,' 
'hough.' Kemble says:^ "Originally a point of land, 
formed like a heel, or boot, and stretching into the plain ; 
perhaps even into the sea." 

Hoylake. — The name of this favourite golfing resort 
has this much in common with that of Liverpool— in both 

' William Mitford, in his Harmony in Language, published in 1S04, 
says (p. 24) : " A fashion has been growing to pronounce the word 
tnerchant (formerly written as spoken, marchant, from the French 
marchanJ) as if it were written murchant. Here, as in some other 
instances, the corruption of orthography has tended to the corruption 
of pronunciation." 

^ O. Nor. a is really a diphthong representing a -f u, pron. on 
or oii). 

3 English .Surnames, 5th ed., 1897, p. 120. 

^ As an instance of what can be done in quite modern times in the 
way of wilfully corrupting land-names, it may be mentioned that in 
Bacon's Geological Map of England and Wales (1891) this island is 
called Elhoroui^h ! 

^ Codex Diplomaticus, iii. xxxi. 


cases the appellation of a piece of water has eventually been 
bestowed upon a village on its banks. As to the Hoyle 
Lake, Ormerod says ■} " The adjacent lake, antiently called 
'Lacus de Hilburgheye,' and ' Heye-pol ' . . . derives its 
present name from two large sandbanks, which afford in 
stormy weather a salutary refuge to the vessels frequenting 
the port of Liverpool . . . the blending of the two lights 
[Bidston and Leasowe] being the signal that the vessel is 
right for Hoyle Lake." Having in view the Leasowe sub- 
marine forest, a wandering philologist might be tempted to 
assert that ' Hoyle ' is the Gael, coill or choill (pron. hoyle), 
'a wood'; but it is no doubt simply the Nor. Eng. hoyle = 
'hole,' 'hollow' (in this case a hollow filled with water), 
' lake ' being added to the name when the signification of 
hoyle had been forgotten. In the South Lancashire dialect 
" long becomes sometimes ow — now for Jio ; sometimes 
oy— hoyle for hole."'- Nodal and Milner,^ and also Holland,^ 
have hoivle but not hoyle in their respective glossaries ; but 
there is no question of the former, if not of the present, exis- 
tence of the form hoyle. 

Irby. — "Hugh Lupus granted this township as 'the 
manor of Erby in Wirhalle,' in his charter to the Abbey of 
St. Werburgh in 1093."^ The name of this Norse l>yr or 
settlement is apparently eponymous. Says Taylor :° " From 
the common Scandinavian name of Ivar or Ingvar we have 
Jurby in the Isle of Man, formerly Ivorby, Irby in Lincoln- 
shire and in Yorkshire." 

Landican. — Domesday spelling, Lafidcchene . As the 
existence of a very ancient church is traced here, Mr. 
Irvine'^ had no difficulty in taking the first element of the 
name to be the common Welsh ecclesiastical Han, and, fol- 
lowing a usual canon, the latter half to be the patronymic 
of a saint, in this case that of a holy person long forgotten. 
But I think we have here, instead of the Welsh llaii, its 

1 Chesh., 1819, ii. 273. 

- Picton, South Lancashire Dialect, Proceedings Lit. and Phil. Soc. 
L'p"t)l, vol. xix. (1S64-65), p. 47. 

3 Lancashire Glossary, Eng. Dial. Soc, 1875-82. 

* Cheshire Glossary, Eng. Dial. Soc, 1S84-86. 

^ Ormerod, Chesh., 1819, ii. 280. 

^ Names and their Histories, 1896, p. 351. 

^ Transactions Hist. Soc. Lane and,, 1891-92, pp. 279 sqq. 


Gaelic equivalent lann, whose plural is /anndaickean, which 
might explain the whole word. 'Landican ' may have had 
the primary meaning of ' the enclosures,' or the later signi- 
fication of 'the churches.' 

Larton. — The old spelling was Lareton, which ap- 
parently means ' the marsh hamlet '— Mdle. Eng. lar, laire, 
'marsh,' 'bog,' 'mire,' from O. Nor. leir (Dan., Norw., 
Swed., Icr). Halliwell's ' Diet, of Archaic and Provincial 
Words '(1850) has lare, 'quagmire,' 'bog.' There are or 
were various cars or mosses in the vicinity of Larton. 

Leasowes, The.—' The Pastures'— A.-Sax. Iceswe, ' a 

Ledsham is clearly eponymous. In Domesday we 
have Levetesham ; and as the names of owners of neighbour- 
ing manors occur Leuvede and Leviett. The place was 
therefore the ham, or home, of this Saxon family, 

Leighton.— Domesday, Lestoite. The numerous Leigh- 
tons and Leytons in England were tihis or farmsteads on 
meadow or pasture land— A.-Sax. hah, ' meadow,' 'pasture,' 
' lea.' The s in the Norman Lestone represents the Saxon 
guttural ; cf. Crosfone = Croughton. 

Liscard. — The meaning of this name is involved in some 
obscurity. Ormerod said^ that the place was formerly 
called Listark ; but this was an error for Liscark.^ Irvine 
remarks^ that the medieval appellations correspond to those 
of Disk card in Cornwall, with which place it is natural to 
compare our Liscard ; and as to the Cornish town I have the 
following note from a respected local antiquary : 

"The ancient name of Liskeard is Lyscerruyt, and 
appears so in the Bodmin Manumission Book, written, it is 
supposed, between a.d. 940 and 1040. It is said to be the 
only original record relating to Liskeard anterior to the 
Norman Conquest which is now extant. Liskeard was for 
many centuries called ' Liskerrett,' otherwise ' Liskeard.' 
The etymology is very uncertain, but the generally received 
opinion is that it is derived from Hys or Lis, ' a manor house 
or fort,' and Caer, 'a walled town.' " 

1 Chesh., 1819, ii. 264. 

2 Helsby's Ormerod's Chesh., 1882, ii. 478. 

3 Transactions Ilisl. Soc. I-anc. and Chesh., 1891-92, pp. -il*^ sqq. 


The latter part of this derivation must at once be dis- 
carded. Lis is perhaps the Cymric llys, ' palace,' ' hall,' 
' court,' and the succeeding portion of the name may easily 
be the patronymic of the ancient proprietor of the estate ; 
but I am rather inclined (at any rate as to our Cheshire 
Liscard) to favour a derivation from the Gael, lis or lios, 
'enclosure,' etc., combined with Gael, ceard (the Scottish 
surname Caird), 'smith,' which would give 'the smith's 
place,' corresponding to the English Smethwicks. This 
etymology would appear to be clinched by a reference to 
Joyce.^ He gives Liscartan as representing ' the fort 
[? place] of the forge,' from ceardcha (pron. cardha), ' a forge ' 
— modern spelling carte, cart, cartaii^ etc. ; ceard (^xor\. card) 
'smith,' 'artificer.' 

Meols. — Domesday Melas — O. Nor, vielr (genit. case, 
inels) ' sandhill,' ' sandbank.' 

MoUington occurs in Domesday as Molintone. The 
English MoUingtons denote estates originally held by the 
Saxon Moiling tribe or family. 

Moreton indicated ' the tun or farmstead on or by the 
moor ' — A. Sax. mor, ' moor.' 

Ness, Nest on. — The Domesday Nesse and Nestone. 
Neston was 'the tun or farmstead on the ness,' Le., head- 
land — A. -Sax. ncess, O. Nor. ties, ' headland,' 'promontory.' 

Netherpool. — ' The ;Lower Pool '— A.-Sax. neother, 

New Brighton. — This is a modern name bestowed in 
imitation of the Sussex watering-place, which was anciently 
called Brihthelmestan, i.e., ' the Stone of Brihthelm.' The 
A.-Sax. stiin, literally 'stone,' 'rock,' sometimes indicated 
a residence or castle built of stone, or on a rock,^ and a 
monumental or boundary stone. A very slight corruption 
of the old name of the Sussex Brighton was current up to 
the early years of the present century. 

1 Irish Local Names Explained, 1884, pp. 66 and 98, 

2 The Dutch and Flemish stem {\,xox\.. stCin), literally 'stone,' also 
denotes 'a castle': the old Scheldt-side keep at Antweip, now converted 
into a " Musceum van Oudhcden " (Museum of Antiquities), is still 
called " Het Steen "—The Stone (Castle). 


Noctorum. — This name has nothing to do with the 
Latin language, despite its genitive plural appearance. 
Domesday has Chenoterie, which, if as usual we turn the 
Norman ch back into k, does not bear so very distant a 
resemblance to the name of the township as it was spelled 
in the thirteenth century, viz., Knocttyrum, or even as 
Ormerod (1819) has it, Knoctorimu The first element of 
these renderings at once suggests the Gael, awe (knock), 
'hill.' For the second portion of the name the Gael. 
druim, ' ridge,' has been hazarded ; but this is obviously 
impossible. We must rather, I think, look for some such 
word as Gael, torran (diminutive of torr), 'grave,' 'tomb.' 
The nearest approach to the name which I can find in 
Joyce's works on Irish topographic etymology is Knockatarry 
(Ir, Cnoc-d! -tairhh\ ' the hill of the buU.'^ 

Overchurch is ' the church on the banks,' i.e., of a 
mere which is now drained — A.-Sax. bfer {f=v), 'shore,' 

Overpool. — ' The Upper Pool ' — A.-Sax. ofer (J= v), 
' upper.' This is the Domesday Pol. 

Oxton. — The name of this place figures in meditTval 
documents as Oxon, which some have erroneously thought 
to be a corruption of a British designation. There is, how- 
ever, no reason why a patient animal, anciently more used 
than to-day, should not give its name to an Anglo-Saxon 
tiin or settlement just as it has given it to a ford (Oxford) 
and to a creek (Oxwich). On the other hand, the name 
may be, and doubtless is, eponymic ; we know that Oxman- 
town, Dublin, was originally ' Ostmentown,' or the dwelling- 
place of the Eastern men or Danes. It has been thought 
that the Oi:;godestun mentioned in the will of Wulfric Spott, 
Earl of Mercia, which was drawn up about 1004 a.d., re- 
ferred to the Wirral Oxton ; but I can see no reason to 
dissent from Dugdale's identification of this Oggodestiin with 
Ogston in Derbyshire.- The Ox in ' Oxton ' is, however, 
probably a corruption of a similar personal name in its 
genitive or possessive case. 

' Irish Local Names Explained, 1884, p. 60. 
" Monasticon Anglicanum, 16S2, i. 266. 


Parkgate. — " The name Parkgate is said to have been 
given by the labourers who were engaged in making the 
sea-wall, and its proximity to Leighton Park originated it ; 
before that it was called New Quay or New Haven. "^ 

Pensby. — In the time of Henry VI. we have Pennesby. 
The name no doubt embodies the patronymic of the 
founder of this Norse by or settlement. In Scandinavian 
peti is the equivalent of ' smart,' ' spruce,' ' nice.' 

Poolton, not mentioned in Domesday, was formerly written 
Pulton and Poulton, i.e. ' the tun or farmstead by the [Wal- 
lasey] Pool ' — A.-Sax. pol. The second Poulton, called in 
Domesday PoJitone [?J, took its name from the Brom- 
borough Pool, the distinguishing suffix Lancelyn being the 
patronymic of ancient owners of the manor. 

Prenton. — In Domesday the name of this township 
figures as Prestune. The numerous Prestons in England 
indicate the tuns or habitations of priests — A.-Sax. /r^t^^A 
' priest.' ' Prenton ' is synonymous with ' Preston.' "Ead- 
berht, the last true-born King of Kent, was surnamed Pren, 
or the Priest, for he had been ordained. "- 

Puddington. — The name of this village, near the Dee, 
has nothing to do with Christmas cheer. In Domesday it 
occurs as Potintone. The English Puddingtons are usually 
considered to be settlements of the Saxon PcTeting tribe. 
In Anglo-Saxon / and d were often used indiscriminately for 
each other. 

Raby. — The Domesday spelling is Rabie. The Lanca- 
shire ' Roby ' was originally identical with this name. In 
both ' Raby ' and ' Roby ' we probably have embodied the 
common animal-name of the earliest Scandinavian owner of 
the by or settlement, the O. Nor. n?', pron. as row = noise — 
Dan.-Norw. ran, Swed. rd, both pron. ra7v — meaning 'a 


1 Mrs. Gamlin, 'Twixt Mersey and Dee, 1897, p. 251. 
" Kemble, Names, Surnames, and Nic-namcs of the Anglo-Saxons, 
Proceedings Archxol. Inst. Gl. Britain, 1S46, p. 93. 


Soui'hall — Domesday has a Saihak, which developed 

later into Sa/g/ia//. This has (perhaps naturally) been 
construed as ' the hall of the sallows or willows ' — A. -Sax. 
sealh, Scot, saugh, 'a sallow,' 'a willow;' but the A.-Sax. 
heaUi is sometimes a word of doubtful meaning. Kemble 
decided that it must represent ' a stone house,' 'a hall,' and 
this canon was long followed. Recent investigations, 
however, show that the word will very often not bear this 
interpretation, and that it rather means ' a slope,' which is 
in accordance with the signification of the O. Nor. hallir), 
' a slope,' ' a hill.'^ On the whole, therefore, we must take 
Saughall and Soughall to represent ' willow-slope.' 

Seacombe. — The Saxons transformed the Wei. cwin, 
'valley,' 'dale,' 'glen,' 'dingle,' into coi/ide. Seacombe is 
close to the point where the Mersey discharges itself into 
the Irish Sea. 

Shotwick. — Both components of this name are of 
somewhat uncertain origin. The Domesday spelling was 
Soto7vic/ie,2ind in the thirteenth century we find Schotewycke, 
v.'hich shows that the early pronunciation was practically as 
it is to-day. The three best-known English place-names 
with the element .f/z^/— Aldershot, Bagshot, and Oakshot — 
being situated on or near the sites of ancient forests, are 
generally presumed to derive their last syllable from a com- 
bination of the genitive or possessive s with the O. Eng. 
/lolf, ' copse,' ' wood ' ; but other names with shot, and 
which have the termination sete in Domesday, are thought 
by some to be referable, as to this, to a doubtful O. Eng. 
seotu, ' common,' ' pasture ' ; while Shottery (near Stratford- 
on-Avon), anciently Scotta-nth, it is concluded, might be 
' trout-beck ' — A.-Sax. scedt or scebta, ' trout. '^ 

' In a description of the district (Hales Owen) where Shcnstone 
resided, which is prefixed to an edition of ihe poet's works published 
in 1793. it is curious to note how often the word 'slope' and its 
derivatives are used by the writer, who probably had not the faintest 
idea that the name Hales was itself significative of the nature of the 
landscape. .Sheriff Hales, also in Shropshire, is called in Domesday 
Halas, ihe plural oi healh, 'a slope,' h after a consonant being dropped 
in the inflection of Anglo-Saxon nouns. 

2 See Taylor, Names and their Histories, 1896, pp. 381 and 385 


Referring to Shotwick, Ormerod says^ : " Here, as in 
most cases in Cheshire where this termination of name 
occurs, were formerly saltworks." This being so, and salt 
deposits having been worked at a very early period in 
Britain, and elsewhere — possibly in Britain by the pre- 
Celtic agglutinative-language (neolithic) race or races — it is 
not improbable that Shotwick was a Saxon ante-Norse 
settlement, and that the wiciji) (usually meaning, ' dwelling,' 
'habitation,' 'village,' 'station') of the Saxons was con- 
verted by the Vikings, i.e.^ the Creekers, into their charac- 
teristic word for ' creek,' ' bay,' namely, v'lk (wick). VV^e seem, 
however, to have fairly distinct evidence that the A. -Sax. 
ivic^ besides its general signification of ' habitation ' (Lat. 
vicus (1) ' quarter or street of a town,' (2) ' village,' ' hamlet '; 
Gr. qIko% — pron. weekos — 'dwelling-place'), also meant 'a 
marsh.' This was Dr. Leo's supposition, and Kemble, 
referring to it, says- : " Some plausibility is given to the 
suggestion, partly by the frequent use of the termination 
-wich in places near salt-pools, and by the occurrence of 
such names as hi'ebdiv'u [Reedwich]." It has been thought, 
again, not inconceivable that Shotwick may be a corruption 
of Scalt-wic or Salt-nnc. A place of this name is mentioned 
in a charter of /Ethelbaid of Mercia, a.d. 716-717,^ and 
elsewhere, but it is identified with Droitwich.'* 

But, on the whole, the most probable explanation of 
' Shotwick' is that which defines it as 'the (salt) station on 
the shot or spit of land ' (perhaps extending into the river 
Dee) — A.-Sax. scedt or scebt, 'corner,' 'division,' 'portion,' 
' tract.' In Wright's ' Diet, of Obsolete and Provincial 
English' (1857) shot=^zx\ angle of land,' and Halliwell's 
' Diet, of Archaic and Provincial Words ' (1850) has shott = 
'field,' 'plot of land.' 'Shot' is apparently a doublet of 
'sheet,' both being ultimately from A. -Sax. scebtan, 'to 
shoot,' 'to extend'; but while 'sheet' takes its pronuncia- 

^ Chesh., 1819, ii. 309. 

^ Codex Diplomaticus, iii. xli. •* Ibid., i. 81. 

•* Had this theory been tenable it would have been necessary to 
explain that the voicing of s as sh in Old English is sporadic ; ' she ' is 
the A.-S. seS. In Gaelic s, in certain defined positions, including 
when preceded or followed by e or i, takca the sound over which the 
Ephraimites came to grief; while in Hungarian s has the sh sound 
pure and simple. 


tion from what the author of the ' History of English 
Sounds ' (Sweet) would call the second gradatory place in 
the eo line, i.e., from the preterite singular {scedt), ' shot ' 
has taken the vowel-sound of the third place, i.e., from the 
past participle {scoten). Analogous words in the Teutonic 
languages are : Ger. schote, Dan.-Norw. skJ4>d, Swed. shot, 
all = ' sheet ' {fiaut.), while Du. schot is ' a partition,' ' a 
division.' The place-component shot has frequently given 
rise to discussion in Notes and Queries, the most useful 
references probably being 6th ser. viii. 369, 412, 523 (1883) ; 
8th ser. i. 148, 214, 337, 419, 484 (1892). See footnote 
under Fazakerley, in the West Derby Section. 

Spital. — This is the same word with hospital, an Old 
French term, meaning 'guest-house,' which came through 
Low Latin from the classic hospitiiim, ' hospitality,' and, by 
metonymy, 'guest-chamber,' 'iim.' In Middle English 
hospital suffered apheresis and became spital or spitel. The 
present signification of ' hospital ' is of comparatively late 

Stanney figured in Domesday as Stand, i.e., ' the stony 
or rocky island ' (or riparian tract)— A.-Sax. stdn, ' stone,' 
' rock ' ; ig { = iy), ' island,' ' low riparian land.' 

Stoke. — Our Stokes and Stocktons seem to have been 
generally stockZidiCdL. places — A.-Sax. stocc, 'stump,' 'stake.' 
Occasionally, however, the name might refer (especially in 
marshy districts) to an erection on stakes or piles. The 
Stokes appear in some respects to compare with our Peels 
or old family fortresses or keeps^ — A.-Sax. /// (Lat. pila), 
' pile,' ' pillar.'- 

Storeton. As this name appears as early as Domesday 
— as Stortone — it is not possible that, as has been suggested, 
we have here the Mdle. Eng. stor, meaning ' store,' ' farm- 
stock,' from O. Fr. estoire. The name doubtless signifies 
' the great tun or farmstead ' — A.-Sax. stor (Scand. stbr"), 
Mdle. Eng. stor and store, 'great,' 'large,' 'strong,' a word 

1 " God save the lady of this pel." — Chaucer, Hous of Fame, iii. 220. 

"^ Skeat, however, in his Glossarial Index to Chaucer, 1894, has 
"/^/=peel, small castle — O. Yx.pel, from Lat. ace. /«/«;«." 

» We meet with the expression stor-thorp (n. pi.), 'large villages,' in 
the Saga known as the Fagrskinna. 


which has not come down to our day in England^ unless it 
still be used in a few remote provincial districts, while it 
flour ishes vigorously in Scandinavia. 

Sutton. — Domesday St^dtone, 'the south tiin or farm- 
stead ' — A. -Sax. si'tih, 'south.' 

Thingwall. — Domesday has Tinguelle. This place was 
the ' Parliament Field ' of the Wirral Norsemen — O. Nor. 
thing, ' parliament,' vollr (dat. sing, velli), ' field.' See 
Thingwall in the West Derby Section. 

Thornton Hough. — Domesday has Torinto7ie ; later 
we have Thor7ieto?i-en-h-Hogh. The English Thorntons 
(A.-Sax. \orn, 'thorn'; /////, 'farmstead') belong to the 
numerous class of place-names derived from large plant-life, 
such as Acton (Oakton), Ashton, Appleton, etc. A hogh or 
hough (A.-Sax. hbJj, ho) is a point of land formed like a heel. 
See HooTON. 

Thurstaston. — The Domesday spelling of this Dano- 
Saxon name is Tiirstaiiefone, which early in the fourteenth 
century had become Thurstaneston. There are two possi- 
bilities here : the village may originally have been ' the tun 
(or farmstead) of Thor's stone '; or the settlement may have 
been named after a proprietor who boasted the same patro- 
nymic as the Norse king Thorstein, who died 874 a.d. The 
former theory is apparently, however, confirmed by the 
existence on Thurstaston Common of a large and remark- 
able stone or rock, which Picton- concluded to be a relic of 
Saxon or Danish heathendom. 

Tranmere does not figure in Domesday. The earliest 
spellings recorded dispose of any apparent connection with 
a mere or marsh. In documents of the thirteenth century 
occur Traninuil, Trafiino/i, and Traninoel ; later still we 
have Tranmore. There does not seem to be any reason to 
doubt the general accuracy of the etymology put forward by 
Mr. Irvine^ — Wei. Tre-yn-moel, ' the hill settlement,' especi- 
ally as this is borne out physiographically. 

^ Yigfusson, however, thinks it is embodied in 'sturdy,' Nor. Eng. 
- Helsby's Ormerod's Chesh., 1SS2, ii. 511. 
' Transactions Hist. Sec. Lane, and Chesh., 1S91-92, pp. 279 sqq. 



Upton. — This name occurs in Domesday as Opsone, and 
is evidently due to the elevated situation of the Anglo-Saxon 
fiin or farmstead. 

Wallasey. — "The angles of the termination of the 
promontory of Wirral consist of two rocky elevations which 
have apparently been separated from the mainland by the 
streams of the Dee and the Mersey at some distant period. 
One of these is called IValea in Domesday, and the other 
Cerchebia, or Kirkby, in a charter of 1081. The latter of 
these was afterwards denominated ' West Kirkby,' as a dis- 
tinction from Kirkby-in-Walley, the name assumed by the 
former parish as early as the thirteenth century, and which 
was shortly afterwards changed to Walayeseghy^ 

At any rate, Wallasey at the present day is almost an 
island, hemmed in as it is by Liverpool Bay, the Mersey, 
Wallasey Pool, and the Birket or Fender. Many local 
antiquaries have definitively accepted as the etymology of the 
name Wallasey the A.-Sax. combination Weal[h)as-ig, ' the 
Welshmen's island,' as it is known that the ancient Britons 
held their ground in the Wirral peninsula until a compara- 
tively late period compared with other Anglo-Saxon and 
Scandinavian districts ; but it appears to have escaped 
observation that upon the coast of Essex, whence the Britons 
were early driven, there is an island with practically the 
same name, viz.^ Wallasea, which Taylor- says is " an island 
surrounded by a se3.-7vall or embankment." This is, I think, 
the best explanation of the name. The A.-Sax 7uea/l or 7c>all 
(genit. wealles or walks ; nom. pi. iveallas or wallas), 'wall,' 
'rampart,' etc.; the Dan.-Norw. val, 'bank,' 'shore'; and 
the Swed. and Ger. wall, Du. ival, 'dam,' 'dike,' 'rampart,' 
' shore,' are from the Lat. valbim, 'rampart.' The suffix ey 
is the A.-Sax. ig ( = ly), O. Nor. ey, of which ea, in the case 
of island-names, is a corruption. We may take it that the 
first two syllables in ' Wallasey ' and ' Wallasea ' are the 
plural of wall, and not the genitive singular, and that there- 
fore the names should be read literally as ' embankments 
island,' not 'embankment island,' there probably having 
been a series of sea-resisting dams on each island, although, 

^ Ormerod's Chesh., 1819, i'. 261. 

' Names and their Histories, 1896, p. 373. 


as to the Wirral island, this was apparently not the case at 
the time of the Domesday Survey, which, as we have seen, 
gives V/aka. It is interesting to note that there exists at the 
present day a Wallasey Embankment Commission, whose 
members are elected triennially. 

West Kirkby.— See under Wallasey. In a charter 
of 1 08 1 the name of this rising watering-place occurs as 
Cerchebi and Cenhebia, the Normanized form of the original 
Norse name — O. Nor. kirkja (Dan.-Norw. kirke, Swed. 
kyrka, Scot, kirk), 'church'; Scand. /{y, 'village.' 'West' 
was prefixed to 'Kirkby' in order to distinguish this church- 
village from the old Wallasey Kirkby. 

Whitby. — The name of this village near Ellesmere Port 
is the Norse equivalent of the EngUsh Whitton, or ' white 
town' — O. Nor. and A.-Sax. hw'it, 'white,' 'bright'; Scand. 
by, ' settlement,' ' village.' The Saxons and Scandinavians 
appear to have used the adjective Jnvit, ' white,' to dis- 
tinguish stone buildings from the usual sombre wooden 
erections. Compare 'Whitchurch.' 

Willaston. — " From circumstances which it would be 
vain to inquire into, the township from which the Hundred 
of Wirral derives its Norman name of IVilaveston has 
escaped notice in the Domesday Survey. It appears first 
... in 1230 . . . [as] Willaston.'''''^ This name without 
doubt embodies the personal appellation (perhaps W'jgldf, 
* War Heritage ') of the original Saxon proprietor of the tun 
or manor. 

Wirral. — The name of the celebrated peninsula occurs 
in the Anglo-Saxon Chronicle'^ as IVirheal, literally (prob- 
ably) 'myrtle-corner' — A.-Sax. tcur, 'a myrtle-tree'; heal, 
' angle,' ' corner,' ' slope '; the supposition being that this 
corner of land was originally overgrown with bog-myrtle. It 
is interesting to note that one of the mediseval spellings of 
the name of the peninsula recurs so late as 1820 in a Crown 
deed conveying the bailiwick of Wirehall to one John 
Williams of Liverpool, to whom the Government sold it in 
fee. Mr. Mortimer has printed the conveyance verbatim? 

' Ormerod's Chesh., 1819, ii. 300. 

^ Under the year 805 — Alfred's Wars with the Danes. 

^ Hist, of Wirral, 1847, p. 154. 


The form IVirhal (" on Wirhalum ") is found in the will of 
Wulfric Spott {ob. loio), as printed by Dugdale.^ Sir Peter 
Leycester has the present spelling Wirral in the seventeenth 

^ Monasticon Anglicanum, 1682, i. 266. 
^ Antiquities of Cheshire, 1673, P- 92- 


{This list docs not represent ail the 'works and records consulted : it 
contains only those to which actual reference is made, and it does not 
include dictionaries and periodicals. ] 

Ancient Battlefields in Lancashire. Hardwick, 1S82. 
Angelsdchsische Ortsnamen (' Rectitudines Singularum Personarum '. 

Leo, 1842. 
Anglo-Saxon a7id Old English Vocabularies. Wright, Wiilcker, 1884. 

, An 

s^lo-Saxon Chronicle 

. Gibson's ed. 

, 1692. 


Thorpe's ed 

, 1861. 


Earle's ed. , 



Plummer's ed., 1889 


glo-Saxon Reader. 

Sweet, 1S94. 

Annates Cambria (' Monumenta Historica Britannica '). Petrie, Sharpe, 

Hardy, 1848. 
Antiquities of Cheshire. Leycester, 1673. 
Beowulf. Thorpe's 3rd ed., 1S89. 
Britannia. Camden, 1586 and 1607 eds. 
do. do. Gibson's ed., 1772. 

do. do. Gough's ed., 1789. 

Brut y Tyioysogion, or Chronicle of the [PFelsh] Princes (' Monumenta 

Historica Britannica'). Petrie, Sharpe, Hardy, 1848. 
Cartularium Saxonicum. Birch, 1885, etc, 
Celtic Britain. Rhys, 1884. 
Chancery Rolls (Rotuli Cancellarii). 
Charter Rolls (Rotuli Chartarum). 
Chaucer's Works. Skeat's ed., 1S94, etc. 

Cheshire Glossary. Holland. English Dialect Society, 18S4-86. 
Close Rolls (Rotuli Litterarum Clausarum). 
Codex Diplomaticus ALvi Saxonici. Kemble. English Historical 

Society, 1839-48. 
Corpus Poeticum Boreale. Vigfusson, Powell, 1S83. 
Danes and Norzvegians in England. Worsre, 1852. 
Diplotnatarium Anglicum ^vi Saxonici. Thorpe, 1865. 
Domesday. Ellis, 1816. 

Domesday Book and Beyond. Maitland, 1897. 
Domesday Cheshire aiid Lancashire. Beamont, 1882. 
Domesday Facsimile of Cheshire and Lancashire. James, 1S61-63. 


Ducky /Records (Ducatus Lancastrise). 

EtJglish Surnames : Their Sources and Significations. Bardsley 

5th ed., 1897. 
English Village Community. Seebohm, 1884. 
Ethekverd's Chronicle (' Rerum Anglicarum Scriptores post Bedam '). 

Savile, 1601. 

do. {'Monumenta Historica Britannica '). Petrie, 

Sharpe, Hardy, 1848. 
Feudal England. Round, 1895. 
Florence of Worcester (Florentius Wigorniensis). Thorpe's ed., 

English Historical Society, 1848-49. 
Geographic Etymology : Dictionary of Place- Names. C. Blackie, 1887. 
Glossariitm Antiquitatum Britannicarum. Baxter, 1733. 
Great de Lacy Inquisition, 1311 ('Three Lancashire Documents'). 

Chetham Society, 1868. 
Greek and Latin Etymology. Peile, 1869. 

Hand' Atlas fiir die Geschichte des Mittelalters. Spruner, Menke, 1880. 
Harmony in Language. Mitford, 1804. 

Historia Ecclesiastica Gentis Anglorum. Bede. Stevenson's ed., 1841. 
do. do. Plummer's ed., 1896. 

Historical English Grammar. Morris, 1875. 
History of Cheshire. Ormerod, 1819. 

do. do. Helsby's ed., 1882. 

History of Christian Names. Miss Yonge, 1884. 
History of Cumberland. Jefferson, 1842. 

History of English Sounds from the Earliest Period. Sweet, 1 888, 
History of Evert on. Syers, 1830. 
History of Lancashire. Baines, 1836. 

do. do. Harland and Herford's ed., 1868-70. 

do. do. Croston's ed., 1888-93. 

History of Leverpool. Enfield, 1774- 
History of Liverpool. Troughton, 1810. 
do. T. Baines, 1852. 

History of Manchester. Whitaker, 1775. 
History of Preston. Hardwick, 1857. 
History of Sankey. Beamont, 1889. 
History of Wirral. Mortimer, 1847. 
Holinsked's Chronicles. Hooker's ed., 1587. 
Icelandic Sagas, etc. , relating to the Settlements and Descents of the 

Northmen on the British Isles. Vigfusson's Text, 1887 ; Dasent's 

Translation, 1894. 
Irish Local Najues Explained. Joyce, 1884. 
Irish Names of Places. 2nd ser. Joyce, 1875. 
Itinerary (Leland). Hearne's 2nd ed., 1744. 
Lancashire and Cheshire, Past and Present. T. Baines, 1867. 
Lancashire and Cheshire Records Preserved in the Public Record Office, 

London. Selby. Lancashire and Cheshire Record Society, 1882-83. 
Lancashire Glossary. Nodal, Milner. English Dialect Society, 

LandndmabSk (' Islendi'nga Sogur'). Kongelige Nordiske Oldskrift 

Selskab (Copenhagen), 1843. 


Liier Vitce Ecclesice Dunelmensis. Stevenson's ed. Surtees Society, 


do. Sweet's ed. Early English Text 

Society, 1885. 
Manipulus Vocabulorum. Levin, 1570. Early English Text Society, 

MedicEval Surnames and their Various Spellings. Grazebrook. Society 

of Antiquaries, 1897. 
Memorials of Liverpool. Picton, 1875. 
Mersey, Ancient and Modern. Blower, 1878. 
Monasticoti Anglica7nun. Dugdale, 1682 ; Ellis's ed. 1817-30. 
Momwtenta Historica Britamiica. Petrie, Sharpe, Hardy, 1848. 
Moore Charters and Docu7)ients J\elating to Liverpool : Report to the 

City Council. Part I. Picton, 1889. 
Names a7id their Histories. Taylor, 1896. 
Names, Surnames, and Nic-Names of the Anglo-Saxons. Kemble. 

Archaeological Institute of Great Britain, 1846. 
Ninth Lnquisitions (Nonarum Inquisitiones). 

Notes on English Etymohgy. Skeat. Philological Society, /^^^'w. 
Oldest English Texts. Sweet. Early English Text Society, 1885. 
Patent Rolls (Rotuli Litterarum Patentium). 
Pipe Rolls (Magni Rotuli Pipce). 
Place-Names in the Himdi-ed of IVirral. Irvine. Lancashire and 

Cheshire Historic Society, 1891-92. 
Polych-onicon Ranulphi Higden Monachi Cesti-ensis, with Trevisa's 

and another Translation. Babbington and Lumby's ed., 1S65-86. 
Portfolio of Fragments relative to the History and Antiquities of Lan- 
cashire. Gregson, 1817; Harland's ed., 1869. 
Post-Mortem Inquisitions (Calendarium Inquisiiionum Post Mortem). 
Pre-Nor7nan Sculptured Sto7ies i7i La7icashire. Browne. Lancashire 

and Cheshire Antiquarian Society, 18S7. 
Promptoriu7/i Pa7~vulo>-wn. Way's ed. Camden Society, 1843-65. 
Proper Na/zies z« Philological a7id Eth7iological E/iquiries. Picton. 

Literary and Philosophical Society, Liverpool, 1865-66. 
Ptoler/iy Elucidated. Rylands, 1893. 
Rise a7td Prog7-ess of the E7tglish Co/n7/ionwealth : A7tglo-Saxon Period. 

Palgrave, 1832. 
Ry77ie7-'s Fcedera. 
Saxonis Gra7n77iatici Gesta Da7ioru77i. Holder's ed., 1886. 

do. First nine books translated and 

edited by Elton and Powell respectively, 1894. 
Saxons in Engla7id. Kemble. 15irch's ed., 1876. 
Scie7ice of Language. Max Miiller, 1862-64. 
ScSp or Glee77ia7is Tale. Thorpe's ed., 1889. 
Scottish La7id-Nar/ies. Maxwell, 1S94. 
South I^ancashire Dialect. Picton. Literary and Philosophical 

Society, Liverpool, 1864. 
Survey of 1320-1346 ('Three Lancashire Documents'). Chetham 

Society, 1868. 
Testa de Nevill. 
Traces of History in the Names of Places. Edmunds, 1S72. 


' Twixt Mersey and Dee. Mrs. Gamlin, 1897. 
Village Comtminity. Gomme, 1890. 
Words and Places. Taylor, 1864, 1873. 

■ Note. — Several treatises quoted are not inserted in the foregoing 
list owdng to their lack of bibliographical interest. 

The Publications of the following Societies are quoted ; 

Archjeological Institute of Great Britain. 

Camden Society. 

Chester Archaeological Society. 

Chetham Society. 

Early English Text Society. 

English Dialect Society, 

English Historical Society. 

Historic Society of Lancashire and Cheshire. 

Kongelige Nordiske Oldskrift Selskab (Copenhagen). 

Lancashire and Cheshire Antiquarian Society. 

Literary and Philosophical Society of Liverpool. 

Numismatic Society. 

Philological Society. 

Pipe Roll Society. 

Record Society of Lancashire and Cheshire. 

Society of Antiquaries. 

Surtees Society. 


Elliot Stock, 62, PaterHosicr Roiu, London 


Los Angeles 
This book is DUE on the last date stamped below. 

Form L9-100m-9.'52(A3105)444 



Harrison - 
Place-nan es of tl 
^he Liver pool 
district , 


AA 000 393 045 o